Compare commits

..

No commits in common. "mainline" and "data_migration" have entirely different histories.

13 changed files with 296 additions and 750 deletions

View file

@ -46,8 +46,8 @@ category: Network
extra-source-files: CHANGELOG.md
common deps
build-depends: base >=4, containers ^>=0.6.0.1, bytestring, utf8-string ^>=1.0.1.1, network ^>=3.1, time, cmdargs ^>= 0.10, cryptonite, memory, async, stm, asn1-encoding, asn1-types, asn1-parse, publicsuffix, network-byte-order, safe, iproute, mtl, random, servant, servant-server, servant-client, warp, text, unordered-containers, hashable, unicode-transforms, http-client, http-types, unbounded-delays, dlist, formatting
ghc-options: -Wall -Wpartial-fields -O2
build-depends: base ^>=4.12.0.0, containers ^>=0.6.0.1, bytestring, utf8-string ^>=1.0.1.1, network ^>=2.8.0.1, time ^>=1.8.0.2, cmdargs ^>= 0.10, cryptonite ^>= 0.25, memory, async, stm, asn1-encoding, asn1-types, asn1-parse, publicsuffix, network-byte-order, safe, iproute, mtl, random, servant, servant-server, servant-client, warp, text, unordered-containers, hashable, unicode-transforms, http-client, http-types
ghc-options: -Wall
@ -58,7 +58,7 @@ library
exposed-modules: Hash2Pub.FediChord, Hash2Pub.FediChordTypes, Hash2Pub.DHTProtocol, Hash2Pub.ASN1Coding, Hash2Pub.ProtocolTypes, Hash2Pub.PostService, Hash2Pub.RingMap
-- Modules included in this library but not exported.
other-modules: Hash2Pub.Utils, Hash2Pub.PostService.API
other-modules: Hash2Pub.Utils
-- LANGUAGE extensions used by modules in this package.
other-extensions: GeneralizedNewtypeDeriving, DataKinds, OverloadedStrings
@ -91,21 +91,7 @@ executable Hash2Pub
-- Base language which the package is written in.
default-language: Haskell2010
ghc-options: -threaded -rtsopts -with-rtsopts=-N
executable Experiment
-- experiment runner
import: deps
build-depends: Hash2Pub
main-is: Experiment.hs
hs-source-dirs: app
default-language: Haskell2010
ghc-options: -threaded
ghc-options: -threaded
test-suite Hash2Pub-test

View file

@ -1,7 +1,7 @@
# Hash2Pub
***This is heavily WIP and does not provide any useful functionality yet***.
I aim for always having the `mainline` branch in a state where it builds and tests pass.
I aim for always having the master branch at a state where it builds and tests pass.
A fully-decentralised relay for global hashtag federation in [ActivityPub](https://activitypub.rocks) based on a distributed hash table.
It allows querying and subscribing to all posts of a certain hashtag and is implemented in Haskell.
@ -10,13 +10,8 @@ This is the practical implementation of the concept presented in the paper [Dece
The ASN.1 module schema used for DHT messages can be found in `FediChord.asn1`.
For further questions and discussins, please refer to the **Hash2Pub topic in [SocialHub](https://socialhub.activitypub.rocks/c/software/hash2pub/48)**.
## Building
The project and its developent environment are built with [Nix](https://nixos.org/nix/).
The development environment can be entered with `nix-shell shell-minimal.nix`. Then the project can be built with `cabal build` from within the environment, or using `nix-shell --command "cabal build" shell-minimal.nix` to do both steps at once.
While the `shell-minimal.nix` environment contains everything necessary for building and testing this project, the `shell.nix` additionally contains the Haskell IDE engine *hie* and the documentation for all used Haskell packages for more convenient development.
Be aware that these need to be build from source and can take a very long time to build.
The development environment can be entered with `nix-shell`. Then the project can be built with `cabal build` from within the environment, or using `nix-shell --command "cabal build"` to do both steps at once.

View file

@ -1,51 +0,0 @@
{-# LANGUAGE OverloadedStrings #-}
module Main where
import Control.Concurrent
import Control.Monad (forM_)
import qualified Data.Text.Lazy as Txt
import qualified Data.Text.Lazy.IO as TxtI
import qualified Network.HTTP.Client as HTTP
import System.Environment (getArgs)
import Hash2Pub.PostService (Hashtag, clientPublishPost)
-- configuration constants
timelineFile = "../simulationData/inputs/generated/timeline_sample.csv"
main :: IO ()
main = do
-- read CLI parameters
speedupStr : _ <- getArgs
-- read and parse timeline schedule
-- relying on lazyness of HaskellIO, hoping it does not introduce too strong delays
postEvents <- parseSchedule <$> TxtI.readFile timelineFile
-- actually schedule and send the post events
executeSchedule (read speedupStr) postEvents
pure ()
parseSchedule :: Txt.Text
-> [(Int, Hashtag, (String, Int))] -- ^ [(delay in microseconds, hashtag, (hostname, port))]
parseSchedule = fmap (parseEntry . Txt.split (== ';')) . Txt.lines
where
parseEntry [delayT, contactT, tag] =
(read $ Txt.unpack delayT, tag, read $ Txt.unpack contactT)
parseEntry entry = error $ "invalid schedule input format: " <> show entry
executeSchedule :: Int -- ^ speedup factor
-> [(Int, Hashtag, (String, Int))] -- ^ [(delay in microseconds, hashtag, (hostname, port))]
-> IO ()
executeSchedule speedup events = do
-- initialise HTTP manager
httpMan <- HTTP.newManager $ HTTP.defaultManagerSettings { HTTP.managerResponseTimeout = HTTP.responseTimeoutMicro 60000000 }
forM_ events $ \(delay, tag, (pubHost, pubPort)) -> do
_ <- forkIO $
clientPublishPost httpMan pubHost pubPort ("foobar #" <> tag)
>>= either putStrLn (const $ pure ())
-- while threadDelay gives only minimum delay guarantees, let's hope the
-- additional delays are negligible
-- otherwise: evaluate usage of https://hackage.haskell.org/package/schedule-0.3.0.0/docs/Data-Schedule.html
threadDelay $ delay `div` speedup

View file

@ -45,36 +45,21 @@ main = do
readConfig :: IO (FediChordConf, ServiceConf)
readConfig = do
confDomainString : ipString : portString : servicePortString : speedupString : remainingArgs <- getArgs
-- allow starting the initial node without bootstrapping info to avoid
-- waiting for timeout
confDomainString : ipString : portString : bootstrapHost : bootstrapPortString : servicePortString : speedup : _ <- getArgs
let
speedup = read speedupString
confBootstrapNodes' = case remainingArgs of
bootstrapHost : bootstrapPortString : _ ->
[(bootstrapHost, read bootstrapPortString)]
_ -> []
fConf = FediChordConf
{ confDomain = confDomainString
, confIP = toHostAddress6 . read $ ipString
, confDhtPort = read portString
, confBootstrapNodes = confBootstrapNodes'
, confStabiliseInterval = 80 * 10^6
, confBootstrapSamplingInterval = 180 * 10^6 `div` speedup
, confMaxLookupCacheAge = 300 / fromIntegral speedup
, confJoinAttemptsInterval = 60 * 10^6 `div` speedup
, confMaxNodeCacheAge = 600 / fromIntegral speedup
, confResponsePurgeAge = 60 / fromIntegral speedup
, confRequestTimeout = 5 * 10^6 `div` speedup
, confRequestRetries = 3
}
sConf = ServiceConf
{ confSubscriptionExpiryTime = 24*3600 / fromIntegral speedup
, confServicePort = read servicePortString
, confServiceHost = confDomainString
, confLogfilePath = "../simulationData/logs/" <> confDomainString <> ".log"
, confSpeedupFactor = speedup
, confStatsEvalDelay = 120 * 10^6 `div` speedup
}
fConf = FediChordConf {
confDomain = confDomainString
, confIP = toHostAddress6 . read $ ipString
, confDhtPort = read portString
, confBootstrapNodes = [(bootstrapHost, read bootstrapPortString)]
--, confStabiliseInterval = 60
, confBootstrapSamplingInterval = 180
, confMaxLookupCacheAge = 300
}
sConf = ServiceConf {
confSubscriptionExpiryTime = fromIntegral $ 2*3600 `div` (read speedup :: Integer)
, confServicePort = read servicePortString
, confServiceHost = confDomainString
}
pure (fConf, sConf)

View file

@ -1,18 +1,26 @@
{
compiler ? "ghc884"
compiler ? "ghc865",
withHIE ? false
}:
let
# pin all-hies for getting the language server
all-hies = fetchTarball {
url = "https://github.com/infinisil/all-hies/tarball/b8fb659620b99b4a393922abaa03a1695e2ca64d";
sha256 = "sha256:0br6wsqpfk1lzz90f7zw439w1ir2p54268qilw9l2pk6yz7ganfx";
};
pkgs = import (
builtins.fetchGit {
name = "nixpkgs-pinned";
url = https://github.com/NixOS/nixpkgs/;
ref = "refs/heads/release-20.09";
rev = "e065200fc90175a8f6e50e76ef10a48786126e1c";
ref = "refs/heads/release-20.03";
rev = "076c67fdea6d0529a568c7d0e0a72e6bc161ecf5";
}) {
# Pass no config for purity
config = {};
overlays = [];
overlays = [
(import all-hies {}).overlay
];
};
hp = pkgs.haskell.packages."${compiler}";
src = pkgs.nix-gitignore.gitignoreSource [] ./.;
@ -30,6 +38,7 @@ in
hlint
stylish-haskell
pkgs.python3Packages.asn1ate
];
]
++ (if withHIE then [ hie ] else []);
};
}

View file

@ -1 +1 @@
(import ./default.nix {}).shell
(import ./default.nix {withHIE = true;}).shell

View file

@ -49,11 +49,9 @@ import Control.Concurrent.STM.TQueue
import Control.Concurrent.STM.TVar
import Control.Exception
import Control.Monad (foldM, forM, forM_, void, when)
import Control.Monad.Except (MonadError (..), runExceptT)
import Control.Monad.IO.Class (MonadIO (..))
import qualified Data.ByteString as BS
import Data.Either (rights)
import Data.Foldable (foldl', foldr', foldrM)
import Data.Foldable (foldl', foldr')
import Data.Functor.Identity
import Data.IP (IPv6, fromHostAddress6,
toHostAddress6)
@ -108,6 +106,9 @@ queryLocalCache ownState nCache lBestNodes targetID
-- the closest succeeding node (like with the p initiated parallel queries
| otherwise = FORWARD $ closestSuccessor `Set.union` closestCachePredecessors (lBestNodes-1) targetID nCache
where
ownID = getNid ownState
preds = predecessors ownState
closestSuccessor :: Set.Set RemoteCacheEntry
closestSuccessor = maybe Set.empty (Set.singleton . toRemoteCacheEntry) $ cacheLookupSucc targetID nCache
@ -258,6 +259,7 @@ handleIncomingRequest :: Service s (RealNodeSTM s)
-> SockAddr -- ^ source address of the request
-> IO ()
handleIncomingRequest nsSTM sendQ msgSet sourceAddr = do
putStrLn $ "handling incoming request: " <> show msgSet
ns <- readTVarIO nsSTM
-- add nodestate to cache
now <- getPOSIXTime
@ -312,6 +314,7 @@ handleIncomingRequest nsSTM sendQ msgSet sourceAddr = do
-- | execute a key ID lookup on local cache and respond with the result
respondQueryID :: LocalNodeStateSTM s -> Set.Set FediChordMessage -> IO (Map.Map Integer BS.ByteString)
respondQueryID nsSTM msgSet = do
putStrLn "responding to a QueryID request"
-- this message cannot be split reasonably, so just
-- consider the first payload
let
@ -432,9 +435,7 @@ respondJoin nsSTM msgSet = do
let
aRequestPart = Set.elemAt 0 msgSet
senderNS = sender aRequestPart
-- if not joined yet, attract responsibility for
-- all keys to make bootstrapping possible
responsibilityLookup = if isJoined nsSnap then queryLocalCache nsSnap cache 1 (getNid senderNS) else FOUND (toRemoteNodeState nsSnap)
responsibilityLookup = queryLocalCache nsSnap cache 1 (getNid senderNS)
thisNodeResponsible (FOUND _) = True
thisNodeResponsible (FORWARD _) = False
-- check whether the joining node falls into our responsibility
@ -488,11 +489,10 @@ requestJoin :: (NodeState a, Service s (RealNodeSTM s)) => a -- ^ cu
requestJoin toJoinOn ownStateSTM = do
ownState <- readTVarIO ownStateSTM
prn <- readTVarIO $ parentRealNode ownState
nodeConf <- nodeConfig <$> readTVarIO (parentRealNode ownState)
let srcAddr = confIP nodeConf
srcAddr <- confIP . nodeConfig <$> readTVarIO (parentRealNode ownState)
bracket (mkSendSocket srcAddr (getDomain toJoinOn) (getDhtPort toJoinOn)) close (\sock -> do
-- extract own state for getting request information
responses <- sendRequestTo (confRequestTimeout nodeConf) (confRequestRetries nodeConf) (\rid -> Request rid (toRemoteNodeState ownState) 1 True Join (Just JoinRequestPayload)) sock
responses <- sendRequestTo 5000 3 (\rid -> Request rid (toRemoteNodeState ownState) 1 True Join (Just JoinRequestPayload)) sock
(cacheInsertQ, joinedState) <- atomically $ do
stateSnap <- readTVar ownStateSTM
let
@ -517,28 +517,28 @@ requestJoin toJoinOn ownStateSTM = do
([], Set.empty, Set.empty)
responses
-- sort, slice and set the accumulated successors and predecessors
-- the contacted node itself is a successor as well and, with few
-- nodes, can be a predecessor as well
newState = setSuccessors (toRemoteNodeState toJoinOn:Set.elems succAccSet) . setPredecessors (toRemoteNodeState toJoinOn:Set.elems predAccSet) $ stateSnap
newState = setSuccessors (Set.elems succAccSet) . setPredecessors (Set.elems predAccSet) $ stateSnap
writeTVar ownStateSTM newState
pure (cacheInsertQ, newState)
-- execute the cache insertions
mapM_ (\f -> f joinedState) cacheInsertQ
if responses == Set.empty
then pure . Left $ "join error: got no response from " <> show (getNid toJoinOn)
else do
-- wait for migration data to be completely received
waitForMigrationFrom (nodeService prn) (getNid toJoinOn)
pure $ Right ownStateSTM
else if null (predecessors joinedState) && null (successors joinedState)
then pure $ Left "join error: no predecessors or successors"
-- successful join
else do
-- wait for migration data to be completely received
waitForMigrationFrom (nodeService prn) (getNid ownState)
pure $ Right ownStateSTM
)
`catch` (\e -> pure . Left $ displayException (e :: IOException))
-- | Send a 'QueryID' 'Request' for getting the node that handles a certain key ID.
requestQueryID :: (MonadIO m, MonadError String m)
=> LocalNodeState s -- ^ NodeState of the querying node
requestQueryID :: LocalNodeState s -- ^ NodeState of the querying node
-> NodeID -- ^ target key ID to look up
-> m RemoteNodeState -- ^ the node responsible for handling that key
-> IO RemoteNodeState -- ^ the node responsible for handling that key
-- 1. do a local lookup for the l closest nodes
-- 2. create l sockets
-- 3. send a message async concurrently to all l nodes
@ -546,23 +546,23 @@ requestQueryID :: (MonadIO m, MonadError String m)
-- 5. repeat until FOUND (problem: new entries not necessarily already in cache, explicitly compare with closer results)
-- TODO: deal with lookup failures
requestQueryID ns targetID = do
firstCacheSnapshot <- liftIO . readTVarIO . nodeCacheSTM $ ns
firstCacheSnapshot <- readTVarIO . nodeCacheSTM $ ns
-- TODO: make maxAttempts configurable
queryIdLookupLoop firstCacheSnapshot ns 50 targetID
-- | like 'requestQueryID, but allows passing of a custom cache, e.g. for joining
queryIdLookupLoop :: (MonadIO m, MonadError String m) => NodeCache -> LocalNodeState s -> Int -> NodeID -> m RemoteNodeState
queryIdLookupLoop :: NodeCache -> LocalNodeState s -> Int -> NodeID -> IO RemoteNodeState
-- return node itself as default fallback value against infinite recursion.
-- TODO: consider using an Either instead of a default value
queryIdLookupLoop _ ns 0 _ = throwError "exhausted maximum lookup attempts"
queryIdLookupLoop _ ns 0 _ = pure $ toRemoteNodeState ns
queryIdLookupLoop cacheSnapshot ns maxAttempts targetID = do
let localResult = queryLocalCache ns cacheSnapshot (lNumBestNodes ns) targetID
-- FOUND can only be returned if targetID is owned by local node
case localResult of
FOUND thisNode -> pure thisNode
FORWARD nodeSet -> do
responseEntries <- liftIO $ sendQueryIdMessages targetID ns Nothing (remoteNode <$> Set.elems nodeSet)
now <- liftIO getPOSIXTime
responseEntries <- sendQueryIdMessages targetID ns Nothing (remoteNode <$> Set.elems nodeSet)
now <- getPOSIXTime
-- check for a FOUND and return it
case responseEntries of
FOUND foundNode -> pure foundNode
@ -585,10 +585,10 @@ sendQueryIdMessages targetID ns lParam targets = do
-- create connected sockets to all query targets and use them for request handling
nodeConf <- nodeConfig <$> readTVarIO (parentRealNode ns)
let srcAddr = confIP nodeConf
srcAddr <- confIP . nodeConfig <$> readTVarIO (parentRealNode ns)
-- ToDo: make attempts and timeout configurable
queryThreads <- mapM (\resultNode -> async $ bracket (mkSendSocket srcAddr (getDomain resultNode) (getDhtPort resultNode)) close (
sendRequestTo (confRequestTimeout nodeConf) (confRequestRetries nodeConf) (lookupMessage targetID ns Nothing)
sendRequestTo 5000 3 (lookupMessage targetID ns Nothing)
)) targets
-- ToDo: process results immediately instead of waiting for the last one to finish, see https://stackoverflow.com/a/38815224/9198613
-- ToDo: exception handling, maybe log them
@ -596,10 +596,8 @@ sendQueryIdMessages targetID ns lParam targets = do
-- insert new cache entries both into global cache as well as return accumulated QueryResponses for further processing
now <- getPOSIXTime
-- collect cache entries from all responses
foldrM (\resp acc -> do
let
responseResult = queryResult <$> payload resp
entrySet = case responseResult of
foldM (\acc resp -> do
let entrySet = case queryResult <$> payload resp of
Just (FOUND result1) -> Set.singleton (RemoteCacheEntry result1 now)
Just (FORWARD resultset) -> resultset
_ -> Set.empty
@ -609,15 +607,10 @@ sendQueryIdMessages targetID ns lParam targets = do
-- return accumulated QueryResult
pure $ case acc of
-- once a FOUND as been encountered, return this as a result
FOUND{} -> acc
FORWARD accSet
| maybe False isFound responseResult -> fromJust responseResult
| otherwise -> FORWARD $ entrySet `Set.union` accSet
isFound@FOUND{} -> isFound
FORWARD accSet -> FORWARD $ entrySet `Set.union` accSet
) (FORWARD Set.empty) responses
where
isFound FOUND{} = True
isFound _ = False
-- | Create a QueryID message to be supplied to 'sendRequestTo'
lookupMessage :: Integral i
@ -636,9 +629,8 @@ requestStabilise :: LocalNodeState s -- ^ sending node
-> RemoteNodeState -- ^ neighbour node to send to
-> IO (Either String ([RemoteNodeState], [RemoteNodeState])) -- ^ (predecessors, successors) of responding node
requestStabilise ns neighbour = do
nodeConf <- nodeConfig <$> readTVarIO (parentRealNode ns)
let srcAddr = confIP nodeConf
responses <- bracket (mkSendSocket srcAddr (getDomain neighbour) (getDhtPort neighbour)) close (fmap Right . sendRequestTo (confRequestTimeout nodeConf) (confRequestRetries nodeConf) (\rid ->
srcAddr <- confIP . nodeConfig <$> readTVarIO (parentRealNode ns)
responses <- bracket (mkSendSocket srcAddr (getDomain neighbour) (getDhtPort neighbour)) close (fmap Right . sendRequestTo 5000 3 (\rid ->
Request {
requestID = rid
, sender = toRemoteNodeState ns
@ -660,7 +652,8 @@ requestStabilise ns neighbour = do
)
([],[]) respSet
-- update successfully responded neighbour in cache
maybe (pure ()) (\p -> queueUpdateVerifieds (Identity $ senderID p) ns) $ headMay (Set.elems respSet)
now <- getPOSIXTime
maybe (pure ()) (\p -> queueAddEntries (Identity $ RemoteCacheEntry (sender p) now) ns) $ headMay (Set.elems respSet)
pure $ if null responsePreds && null responseSuccs
then Left "no neighbours returned"
else Right (responsePreds, responseSuccs)
@ -675,14 +668,13 @@ requestLeave :: LocalNodeState s
-> RemoteNodeState -- target node
-> IO (Either String ()) -- error or success
requestLeave ns doMigration target = do
nodeConf <- nodeConfig <$> readTVarIO (parentRealNode ns)
let srcAddr = confIP nodeConf
leavePayload = LeaveRequestPayload {
srcAddr <- confIP . nodeConfig <$> readTVarIO (parentRealNode ns)
let leavePayload = LeaveRequestPayload {
leaveSuccessors = successors ns
, leavePredecessors = predecessors ns
, leaveDoMigration = doMigration
}
responses <- bracket (mkSendSocket srcAddr (getDomain target) (getDhtPort target)) close (fmap Right . sendRequestTo (confRequestTimeout nodeConf) (confRequestRetries nodeConf) (\rid ->
responses <- bracket (mkSendSocket srcAddr (getDomain target) (getDhtPort target)) close (fmap Right . sendRequestTo 5000 3 (\rid ->
Request {
requestID = rid
, sender = toRemoteNodeState ns
@ -704,11 +696,10 @@ requestPing :: LocalNodeState s -- ^ sending node
-> RemoteNodeState -- ^ node to be PINGed
-> IO (Either String [RemoteNodeState]) -- ^ all active vServers of the pinged node
requestPing ns target = do
nodeConf <- nodeConfig <$> readTVarIO (parentRealNode ns)
let srcAddr = confIP nodeConf
srcAddr <- confIP . nodeConfig <$> readTVarIO (parentRealNode ns)
responses <- bracket (mkSendSocket srcAddr (getDomain target) (getDhtPort target)) close
(\sock -> do
resp <- sendRequestTo (confRequestTimeout nodeConf) (confRequestRetries nodeConf) (\rid ->
resp <- sendRequestTo 5000 3 (\rid ->
Request {
requestID = rid
, sender = toRemoteNodeState ns
@ -744,9 +735,10 @@ requestPing ns target = do
) responses
-- | Generic function for sending a request over a connected socket and collecting the response.
-- Serialises the message and tries to deliver its parts for a number of attempts within a specified timeout.
sendRequestTo :: Int -- ^ timeout in milliseconds
sendRequestTo :: Int -- ^ timeout in seconds
-> Int -- ^ number of retries
-> (Integer -> FediChordMessage) -- ^ the message to be sent, still needing a requestID
-> Socket -- ^ connected socket to use for sending
@ -757,10 +749,11 @@ sendRequestTo timeoutMillis numAttempts msgIncomplete sock = do
let
msgComplete = msgIncomplete randomID
requests = serialiseMessage sendMessageSize msgComplete
putStrLn $ "sending request message " <> show msgComplete
-- create a queue for passing received response messages back, even after a timeout
responseQ <- newTBQueueIO $ 2*maximumParts -- keep room for duplicate packets
-- start sendAndAck with timeout
_ <- attempts numAttempts . timeout (timeoutMillis*1000) $ sendAndAck responseQ sock requests
attempts numAttempts . timeout timeoutMillis $ sendAndAck responseQ sock requests
-- after timeout, check received responses, delete them from unacked message set/ map and rerun senAndAck with that if necessary.
recvdParts <- atomically $ flushTBQueue responseQ
pure $ Set.fromList recvdParts
@ -769,20 +762,19 @@ sendRequestTo timeoutMillis numAttempts msgIncomplete sock = do
-> Socket -- ^ the socket used for sending and receiving for this particular remote node
-> Map.Map Integer BS.ByteString -- ^ the remaining unacked request parts
-> IO ()
sendAndAck responseQueue sock' remainingSends = do
sendMany sock' $ Map.elems remainingSends
sendAndAck responseQueue sock remainingSends = do
sendMany sock $ Map.elems remainingSends
-- if all requests have been acked/ responded to, return prematurely
recvLoop sock' responseQueue remainingSends Set.empty Nothing
recvLoop :: Socket
-> TBQueue FediChordMessage -- ^ the queue for putting in the received responses
recvLoop responseQueue remainingSends Set.empty Nothing
recvLoop :: TBQueue FediChordMessage -- ^ the queue for putting in the received responses
-> Map.Map Integer BS.ByteString -- ^ the remaining unacked request parts
-> Set.Set Integer -- ^ already received response part numbers
-> Maybe Integer -- ^ total number of response parts if already known
-> Maybe Integer -- ^ total number of response parts if already known
-> IO ()
recvLoop sock' responseQueue remainingSends' receivedPartNums totalParts = do
recvLoop responseQueue remainingSends' receivedPartNums totalParts = do
-- 65535 is maximum length of UDP packets, as long as
-- no IPv6 jumbograms are used
response <- deserialiseMessage <$> recv sock' 65535
response <- deserialiseMessage <$> recv sock 65535
case response of
Right msg@Response{} -> do
atomically $ writeTBQueue responseQueue msg
@ -790,12 +782,11 @@ sendRequestTo timeoutMillis numAttempts msgIncomplete sock = do
newTotalParts = if isFinalPart msg then Just (part msg) else totalParts
newRemaining = Map.delete (part msg) remainingSends'
newReceivedParts = Set.insert (part msg) receivedPartNums
if Map.null newRemaining && maybe False (\p -> Set.size newReceivedParts == fromIntegral p) newTotalParts
if Map.null newRemaining && maybe False (\p -> Set.size receivedPartNums == fromIntegral p) newTotalParts
then pure ()
else recvLoop sock' responseQueue newRemaining newReceivedParts newTotalParts
else recvLoop responseQueue newRemaining receivedPartNums newTotalParts
-- drop errors and invalid messages
Right Request{} -> pure () -- expecting a response, not a request
Left _ -> recvLoop sock' responseQueue remainingSends' receivedPartNums totalParts
Left _ -> recvLoop responseQueue remainingSends' receivedPartNums totalParts
-- | enqueue a list of RemoteCacheEntries to be added to the global NodeCache
@ -821,18 +812,6 @@ queueDeleteEntry :: NodeID
-> IO ()
queueDeleteEntry toDelete = queueDeleteEntries $ Identity toDelete
-- | enqueue the timestamp update and verification marking of an entry in the
-- global 'NodeCache'.
queueUpdateVerifieds :: Foldable c
=> c NodeID
-> LocalNodeState s
-> IO ()
queueUpdateVerifieds nIds ns = do
now <- getPOSIXTime
forM_ nIds $ \nid' -> atomically $ writeTQueue (cacheWriteQueue ns) $
markCacheEntryAsVerified (Just now) nid'
-- | retry an IO action at most *i* times until it delivers a result
attempts :: Int -- ^ number of retries *i*
-> IO (Maybe a) -- ^ action to retry
@ -865,7 +844,7 @@ mkServerSocket ip port = do
sockAddr <- addrAddress <$> resolve (Just $ show . fromHostAddress6 $ ip) (Just port)
sock <- socket AF_INET6 Datagram defaultProtocol
setSocketOption sock IPv6Only 1
bind sock sockAddr `catch` (\e -> putStrLn $ "Caught exception while bind " <> show sock <> " " <> show sockAddr <> ": " <> show (e :: SomeException))
bind sock sockAddr
pure sock
-- | create a UDP datagram socket, connected to a destination.
@ -881,6 +860,6 @@ mkSendSocket srcIp dest destPort = do
setSocketOption sendSock IPv6Only 1
-- bind to the configured local IP to make sure that outgoing packets are sent from
-- this source address
bind sendSock srcAddr `catch` (\e -> putStrLn $ "Caught exception while mkSendSocket bind " <> show sendSock <> " " <> show srcAddr <> ": " <> show (e :: SomeException))
bind sendSock srcAddr
connect sendSock destAddr
pure sendSock

View file

@ -166,7 +166,6 @@ fediChordBootstrapJoin nsSTM bootstrapNode = do
currentlyResponsible <- liftEither lookupResp
liftIO . putStrLn $ "Trying to join on " <> show (getNid currentlyResponsible)
-- 2. then send a join to the currently responsible node
liftIO $ putStrLn "send a bootstrap Join"
joinResult <- liftIO $ requestJoin currentlyResponsible nsSTM
liftEither joinResult
@ -199,7 +198,7 @@ convergenceSampleThread nsSTM = forever $ do
-- unjoined node: try joining through all bootstrapping nodes
else tryBootstrapJoining nsSTM >> pure ()
let delaySecs = confBootstrapSamplingInterval . nodeConfig $ parentNode
threadDelay delaySecs
threadDelay $ delaySecs * 10^6
-- | Try joining the DHT through any of the bootstrapping nodes until it succeeds.
@ -223,11 +222,10 @@ tryBootstrapJoining nsSTM = do
bootstrapQueryId :: LocalNodeStateSTM s -> (String, PortNumber) -> NodeID -> IO (Either String RemoteNodeState)
bootstrapQueryId nsSTM (bootstrapHost, bootstrapPort) targetID = do
ns <- readTVarIO nsSTM
nodeConf <- nodeConfig <$> readTVarIO (parentRealNode ns)
let srcAddr = confIP nodeConf
srcAddr <- confIP . nodeConfig <$> readTVarIO (parentRealNode ns)
bootstrapResponse <- bracket (mkSendSocket srcAddr bootstrapHost bootstrapPort) close (
-- Initialise an empty cache only with the responses from a bootstrapping node
fmap Right . sendRequestTo (confRequestTimeout nodeConf) (confRequestRetries nodeConf) (lookupMessage targetID ns Nothing)
fmap Right . sendRequestTo 5000 3 (lookupMessage targetID ns Nothing)
)
`catch` (\e -> pure . Left $ "Error at bootstrap QueryId: " <> displayException (e :: IOException))
@ -246,24 +244,26 @@ bootstrapQueryId nsSTM (bootstrapHost, bootstrapPort) targetID = do
Just (FORWARD resultset) -> foldr' (addCacheEntryPure now) cacheAcc resultset
)
initCache resp
currentlyResponsible <- runExceptT $ queryIdLookupLoop bootstrapCache ns 50 $ getNid ns
pure currentlyResponsible
currentlyResponsible <- queryIdLookupLoop bootstrapCache ns 50 $ getNid ns
pure $ Right currentlyResponsible
-- | join a node to the DHT using the global node cache
-- node's position.
fediChordVserverJoin :: (MonadError String m, MonadIO m, Service s (RealNodeSTM s))
fediChordVserverJoin :: Service s (RealNodeSTM s)
=> LocalNodeStateSTM s -- ^ the local 'NodeState'
-> m (LocalNodeStateSTM s) -- ^ the joined 'NodeState' after a
-> IO (Either String (LocalNodeStateSTM s)) -- ^ the joined 'NodeState' after a
-- successful join, otherwise an error message
fediChordVserverJoin nsSTM = do
ns <- liftIO $ readTVarIO nsSTM
ns <- readTVarIO nsSTM
-- 1. get routed to the currently responsible node
currentlyResponsible <- requestQueryID ns $ getNid ns
liftIO . putStrLn $ "Trying to join on " <> show (getNid currentlyResponsible)
putStrLn $ "Trying to join on " <> show (getNid currentlyResponsible)
-- 2. then send a join to the currently responsible node
joinResult <- liftIO $ requestJoin currentlyResponsible nsSTM
liftEither joinResult
joinResult <- requestJoin currentlyResponsible nsSTM
case joinResult of
Left err -> pure . Left $ "Error joining on " <> err
Right joinedNS -> pure . Right $ joinedNS
fediChordVserverLeave :: (MonadError String m, MonadIO m, Service s (RealNodeSTM s)) => LocalNodeState s -> m ()
fediChordVserverLeave ns = do
@ -311,23 +311,23 @@ joinOnNewEntriesThread nsSTM = loop
where
loop = do
nsSnap <- readTVarIO nsSTM
(lookupResult, parentNode) <- atomically $ do
(lookupResult, cache) <- atomically $ do
cache <- readTVar $ nodeCacheSTM nsSnap
parentNode <- readTVar $ parentRealNode nsSnap
case queryLocalCache nsSnap cache 1 (getNid nsSnap) of
-- empty cache, block until cache changes and then retry
(FORWARD s) | Set.null s -> retry
result -> pure (result, parentNode)
result -> pure (result, cache)
case lookupResult of
-- already joined
FOUND _ ->
pure ()
-- otherwise try joining
FORWARD _ -> do
joinResult <- runExceptT $ fediChordVserverJoin nsSTM
joinResult <- fediChordVserverJoin nsSTM
either
-- on join failure, sleep and retry
(const $ threadDelay (confJoinAttemptsInterval . nodeConfig $ parentNode) >> loop)
-- TODO: make delay configurable
(const $ threadDelay (30 * 10^6) >> loop)
(const $ pure ())
joinResult
@ -342,15 +342,20 @@ nodeCacheWriter nsSTM =
modifyTVar' (nodeCacheSTM ns) cacheModifier
-- TODO: make max entry age configurable
maxEntryAge :: POSIXTime
maxEntryAge = 600
-- | Periodically iterate through cache, clean up expired entries and verify unverified ones
nodeCacheVerifyThread :: LocalNodeStateSTM s -> IO ()
nodeCacheVerifyThread nsSTM = forever $ do
putStrLn "cache verify run: begin"
-- get cache
(ns, cache, maxEntryAge) <- atomically $ do
(ns, cache) <- atomically $ do
ns <- readTVar nsSTM
cache <- readTVar $ nodeCacheSTM ns
maxEntryAge <- confMaxNodeCacheAge . nodeConfig <$> readTVar (parentRealNode ns)
pure (ns, cache, maxEntryAge)
pure (ns, cache)
-- iterate entries:
-- for avoiding too many time syscalls, get current time before iterating.
now <- getPOSIXTime
@ -397,7 +402,8 @@ nodeCacheVerifyThread nsSTM = forever $ do
forkIO $ sendQueryIdMessages targetID latestNs (Just (1 + jEntriesPerSlice latestNs)) (nodesToQuery targetID) >> pure () -- ask for 1 entry more than j because of querying the middle
)
threadDelay $ fromEnum (maxEntryAge / 20) `div` 10^6 -- convert from pico to milliseconds
putStrLn "cache verify run: end"
threadDelay $ 10^6 * round maxEntryAge `div` 20
-- | Checks the invariant of at least @jEntries@ per cache slice.
@ -463,6 +469,7 @@ stabiliseThread :: Service s (RealNodeSTM s) => LocalNodeStateSTM s -> IO ()
stabiliseThread nsSTM = forever $ do
oldNs <- readTVarIO nsSTM
putStrLn "stabilise run: begin"
-- iterate through the same snapshot, collect potential new neighbours
-- and nodes to be deleted, and modify these changes only at the end of
@ -497,26 +504,18 @@ stabiliseThread nsSTM = forever $ do
-- try looking up additional neighbours if list too short
forM_ [(length $ predecessors updatedNs)..(kNeighbours updatedNs)] (\_ -> do
ns' <- readTVarIO nsSTM
nextEntry <- runExceptT . requestQueryID ns' $ pred . getNid $ lastDef (toRemoteNodeState ns') (predecessors ns')
either
(const $ pure ())
(\entry -> atomically $ do
latestNs <- readTVar nsSTM
writeTVar nsSTM $ addPredecessors [entry] latestNs
)
nextEntry
nextEntry <- requestQueryID ns' $ pred . getNid $ lastDef (toRemoteNodeState ns') (predecessors ns')
atomically $ do
latestNs <- readTVar nsSTM
writeTVar nsSTM $ addPredecessors [nextEntry] latestNs
)
forM_ [(length $ successors updatedNs)..(kNeighbours updatedNs)] (\_ -> do
ns' <- readTVarIO nsSTM
nextEntry <- runExceptT . requestQueryID ns' $ succ . getNid $ lastDef (toRemoteNodeState ns') (successors ns')
either
(const $ pure ())
(\entry -> atomically $ do
latestNs <- readTVar nsSTM
writeTVar nsSTM $ addSuccessors [entry] latestNs
)
nextEntry
nextEntry <- requestQueryID ns' $ succ . getNid $ lastDef (toRemoteNodeState ns') (successors ns')
atomically $ do
latestNs <- readTVar nsSTM
writeTVar nsSTM $ addSuccessors [nextEntry] latestNs
)
newNs <- readTVarIO nsSTM
@ -541,8 +540,9 @@ stabiliseThread nsSTM = forever $ do
)
newPredecessor
stabiliseDelay <- confStabiliseInterval . nodeConfig <$> readTVarIO (parentRealNode newNs)
threadDelay stabiliseDelay
putStrLn "stabilise run: end"
-- TODO: make delay configurable
threadDelay (60 * 10^6)
where
-- | send a stabilise request to the n-th neighbour
-- (specified by the provided getter function) and on failure retry
@ -629,15 +629,19 @@ type RequestMap = Map.Map (SockAddr, Integer) RequestMapEntry
data RequestMapEntry = RequestMapEntry (Set.Set FediChordMessage) (Maybe Integer)
POSIXTime
-- TODO: make purge age configurable
-- | periodically clean up old request parts
responsePurgeAge :: POSIXTime
responsePurgeAge = 60 -- seconds
requestMapPurge :: POSIXTime -> MVar RequestMap -> IO ()
requestMapPurge purgeAge mapVar = forever $ do
requestMapPurge :: MVar RequestMap -> IO ()
requestMapPurge mapVar = forever $ do
rMapState <- takeMVar mapVar
now <- getPOSIXTime
putMVar mapVar $ Map.filter (\(RequestMapEntry _ _ ts) ->
now - ts < purgeAge
putMVar mapVar $ Map.filter (\entry@(RequestMapEntry _ _ ts) ->
now - ts < responsePurgeAge
) rMapState
threadDelay $ (fromEnum purgeAge * 2) `div` 10^6
threadDelay $ round responsePurgeAge * 2 * 10^6
-- | Wait for messages, deserialise them, manage parts and acknowledgement status,
@ -652,13 +656,12 @@ fediMessageHandler sendQ recvQ nsSTM = do
-- not change.
-- Other functions are passed the nsSTM reference and thus can get the latest state.
nsSnap <- readTVarIO nsSTM
nodeConf <- nodeConfig <$> readTVarIO (parentRealNode nsSnap)
-- handling multipart messages:
-- Request parts can be insert into a map (key: (sender IP against spoofing, request ID), value: timestamp + set of message parts, handle all of them when size of set == parts) before being handled. This map needs to be purged periodically by a separate thread and can be protected by an MVar for fairness.
requestMap <- newMVar (Map.empty :: RequestMap)
-- run receive loop and requestMapPurge concurrently, so that an exception makes
-- both of them fail
concurrently_ (requestMapPurge (confResponsePurgeAge nodeConf) requestMap) $ forever $ do
concurrently_ (requestMapPurge requestMap) $ forever $ do
-- wait for incoming messages
(rawMsg, sourceAddr) <- atomically $ readTQueue recvQ
let aMsg = deserialiseMessage rawMsg
@ -713,7 +716,7 @@ fediMessageHandler sendQ recvQ nsSTM = do
instance DHT (RealNodeSTM s) where
lookupKey nodeSTM keystring = getKeyResponsibility nodeSTM $ genKeyID keystring
forceLookupKey nodeSTM keystring = (putStrLn $ "forced responsibility lookup of #" <> keystring) >> (updateLookupCache nodeSTM $ genKeyID keystring)
forceLookupKey nodeSTM keystring = updateLookupCache nodeSTM $ genKeyID keystring
-- potential better implementation: put all neighbours of all vservers and the vservers on a ringMap, look the key up and see whether it results in a LocalNodeState
isResponsibleFor nodeSTM key = do
node <- readTVarIO nodeSTM
@ -754,7 +757,7 @@ getKeyResponsibility nodeSTM lookupKey = do
-- new entry.
-- If no vserver is active in the DHT, 'Nothing' is returned.
updateLookupCache :: RealNodeSTM s -> NodeID -> IO (Maybe (String, PortNumber))
updateLookupCache nodeSTM keyToLookup = do
updateLookupCache nodeSTM lookupKey = do
(node, lookupSource) <- atomically $ do
node <- readTVar nodeSTM
let firstVs = headMay (vservers node)
@ -764,25 +767,18 @@ updateLookupCache nodeSTM keyToLookup = do
pure (node, lookupSource)
maybe (do
-- if no local node available, delete cache entry and return Nothing
atomically $ modifyTVar' (lookupCacheSTM node) $ Map.delete keyToLookup
atomically $ modifyTVar' (lookupCacheSTM node) $ Map.delete lookupKey
pure Nothing
)
(\n -> do
-- start a lookup from the node, update the cache with the lookup result and return it
-- TODO: better retry management, because having no vserver joined yet should
-- be treated differently than other reasons for not getting a result.
newResponsible <- runExceptT $ requestQueryID n keyToLookup
either
(const $ pure Nothing)
(\result -> do
let newEntry = (getDomain result, getServicePort result)
now <- getPOSIXTime
-- atomic update against lost updates
atomically $ modifyTVar' (lookupCacheSTM node) $
Map.insert keyToLookup (CacheEntry False newEntry now)
pure $ Just newEntry
)
newResponsible
newResponsible <- requestQueryID n lookupKey
let newEntry = (getDomain newResponsible, getServicePort newResponsible)
now <- getPOSIXTime
-- atomic update against lost updates
atomically $ modifyTVar' (lookupCacheSTM node) $
Map.insert lookupKey (CacheEntry False newEntry now)
pure $ Just newEntry
) lookupSource
@ -797,4 +793,4 @@ lookupCacheCleanup nodeSTM = do
now - ts < confMaxLookupCacheAge (nodeConfig node)
)
)
threadDelay $ fromEnum (2 * confMaxLookupCacheAge (nodeConfig node)) `div` 10^6
threadDelay $ round (confMaxLookupCacheAge $ nodeConfig node) * (10^5)

View file

@ -411,22 +411,10 @@ data FediChordConf = FediChordConf
-- ^ listening port for the FediChord DHT
, confBootstrapNodes :: [(String, PortNumber)]
-- ^ list of potential bootstrapping nodes
, confStabiliseInterval :: Int
-- ^ pause between stabilise runs, in milliseconds
, confBootstrapSamplingInterval :: Int
-- ^ pause between sampling the own ID through bootstrap nodes, in milliseconds
-- ^ pause between sampling the own ID through bootstrap nodes, in seconds
, confMaxLookupCacheAge :: POSIXTime
-- ^ maximum age of key lookup cache entries in seconds
, confJoinAttemptsInterval :: Int
-- ^ interval between join attempts on newly learned nodes, in milliseconds
, confMaxNodeCacheAge :: POSIXTime
-- ^ maximum age of entries in the node cache, in milliseconds
, confResponsePurgeAge :: POSIXTime
-- ^ maximum age of message parts in response part cache, in seconds
, confRequestTimeout :: Int
-- ^ how long to wait until response has arrived, in milliseconds
, confRequestRetries :: Int
-- ^ how often re-sending a timed-out request can be retried
-- ^ maximum age of lookup cache entries in seconds
}
deriving (Show, Eq)
@ -457,12 +445,6 @@ data ServiceConf = ServiceConf
-- ^ listening port for service
, confServiceHost :: String
-- ^ hostname of service
, confLogfilePath :: String
-- ^ where to store the (measurement) log file
, confStatsEvalDelay :: Int
-- ^ delay between statistic rate measurement samplings, in microseconds
, confSpeedupFactor :: Int
-- While the speedup factor needs to be already included in all
}
class DHT d where

View file

@ -10,43 +10,41 @@ module Hash2Pub.PostService where
import Control.Concurrent
import Control.Concurrent.Async
import Control.Concurrent.MVar
import Control.Concurrent.STM
import Control.Exception (Exception (..), try)
import Control.Monad (foldM, forM, forM_, forever, unless,
void, when)
import Control.Monad.IO.Class (liftIO)
import Control.Concurrent.STM.TChan
import Control.Concurrent.STM.TChan
import Control.Concurrent.STM.TQueue
import Control.Concurrent.STM.TVar
import Control.Exception (Exception (..), try)
import Control.Monad (foldM, forM, forM_, forever)
import Control.Monad.IO.Class (liftIO)
import Control.Monad.STM
import Data.Bifunctor
import qualified Data.ByteString.Lazy.UTF8 as BSUL
import qualified Data.ByteString.UTF8 as BSU
import qualified Data.DList as D
import Data.Either (lefts, rights)
import qualified Data.HashMap.Strict as HMap
import qualified Data.HashSet as HSet
import Data.Maybe (fromJust, isJust)
import Data.String (fromString)
import Data.Text.Lazy (Text)
import qualified Data.Text.Lazy as Txt
import qualified Data.Text.Lazy.IO as TxtI
import Data.Text.Normalize (NormalizationMode (NFC), normalize)
import qualified Data.ByteString.Lazy.UTF8 as BSUL
import qualified Data.ByteString.UTF8 as BSU
import qualified Data.HashMap.Strict as HMap
import qualified Data.HashSet as HSet
import Data.Maybe (fromMaybe, isJust)
import Data.String (fromString)
import qualified Data.Text.Lazy as Txt
import Data.Text.Normalize (NormalizationMode (NFC),
normalize)
import Data.Time.Clock.POSIX
import Data.Typeable (Typeable)
import qualified Network.HTTP.Client as HTTP
import qualified Network.HTTP.Types as HTTPT
import System.IO
import Data.Typeable (Typeable)
import qualified Network.HTTP.Client as HTTP
import qualified Network.HTTP.Types as HTTPT
import System.Random
import Text.Read (readEither)
import Text.Read (readEither)
import Formatting (fixed, format, int, (%))
import qualified Network.Wai.Handler.Warp as Warp
import qualified Network.Wai.Handler.Warp as Warp
import Servant
import Servant.Client
import Servant.Server
import Hash2Pub.FediChordTypes
import Hash2Pub.PostService.API
import Hash2Pub.RingMap
import Hash2Pub.Utils
import Debug.Trace
data PostService d = PostService
{ serviceConf :: ServiceConf
@ -57,22 +55,19 @@ data PostService d = PostService
-- ^ for each tag store the subscribers + their queue
, ownSubscriptions :: TVar (HMap.HashMap NodeID POSIXTime)
-- ^ tags subscribed by the own node have an assigned lease time
, ownPosts :: TVar (HSet.HashSet Txt.Text)
-- ^ just store the existence of posts for saving memory,
, relayInQueue :: TQueue (Hashtag, PostID, PostContent)
-- ^ Queue for processing incoming posts of own instance asynchronously
, postFetchQueue :: TQueue PostID
-- ^ queue of posts to be fetched
, migrationsInProgress :: TVar (HMap.HashMap NodeID (MVar ()))
, httpMan :: HTTP.Manager
, statsQueue :: TQueue StatsEvent
, loadStats :: TVar RelayStats
-- ^ current load stats, replaced periodically
, logFileHandle :: Handle
}
deriving (Typeable)
type Hashtag = Text
type PostID = Text
type PostContent = Text
type Hashtag = Txt.Text
type PostID = Txt.Text
type PostContent = Txt.Text
-- | For each handled tag, store its subscribers and provide a
-- broadcast 'TChan' for enqueuing posts
type RelayTags = RingMap NodeID (TagSubscribersSTM, TChan PostID, Hashtag)
@ -90,38 +85,26 @@ instance DHT d => Service PostService d where
threadVar <- newTVarIO =<< myThreadId -- own thread ID as placeholder
subscriberVar <- newTVarIO emptyRMap
ownSubsVar <- newTVarIO HMap.empty
--ownPostVar <- newTVarIO HSet.empty
ownPostVar <- newTVarIO HSet.empty
relayInQueue' <- newTQueueIO
postFetchQueue' <- newTQueueIO
migrationsInProgress' <- newTVarIO HMap.empty
httpMan' <- HTTP.newManager HTTP.defaultManagerSettings
statsQueue' <- newTQueueIO
loadStats' <- newTVarIO emptyStats
loggingFile <- openFile (confLogfilePath conf) WriteMode
hSetBuffering loggingFile LineBuffering
let
thisService = PostService
{ serviceConf = conf
thisService = PostService {
serviceConf = conf
, baseDHT = dht
, serviceThread = threadVar
, subscribers = subscriberVar
, ownSubscriptions = ownSubsVar
--, ownPosts = ownPostVar
, ownPosts = ownPostVar
, relayInQueue = relayInQueue'
, postFetchQueue = postFetchQueue'
, migrationsInProgress = migrationsInProgress'
, httpMan = httpMan'
, statsQueue = statsQueue'
, loadStats = loadStats'
, logFileHandle = loggingFile
}
}
port' = fromIntegral (confServicePort conf)
warpSettings = Warp.setPort port' . Warp.setHost (fromString . confServiceHost $ conf) $ Warp.defaultSettings
-- log a start message, this also truncates existing files
TxtI.hPutStrLn loggingFile $ Txt.unlines
[ "# Starting mock relay implementation"
, "#time stamp ; relay receive rate ;relay delivery rate ;instance publish rate ;instance fetch rate ;total subscriptions"
]
-- Run 'concurrently_' from another thread to be able to return the
-- 'PostService'.
-- Terminating that parent thread will make all child threads terminate as well.
@ -129,11 +112,7 @@ instance DHT d => Service PostService d where
concurrently_
-- web server
(Warp.runSettings warpSettings $ postServiceApplication thisService)
$ concurrently
-- background processing workers
(launchWorkerThreads thisService)
-- statistics/ measurements
(launchStatsThreads thisService)
(processIncomingPosts thisService)
-- update thread ID after fork
atomically $ writeTVar threadVar servThreadID
pure thisService
@ -159,13 +138,38 @@ postServiceApplication :: DHT d => PostService d -> Application
postServiceApplication serv = serve exposedPostServiceAPI $ postServer serv
-- | needed for guiding type inference
exposedPostServiceAPI :: Proxy PostServiceAPI
exposedPostServiceAPI = Proxy
-- ========= constants ===========
placeholderPost :: Text
placeholderPost :: Txt.Text
placeholderPost = Txt.take 5120 . Txt.repeat $ 'O' -- size 5KiB
-- ========= HTTP API and handlers =============
type PostServiceAPI = "relay" :> "inbox" :> Capture "hashtag" Txt.Text :> ReqBody '[PlainText] Txt.Text :> PutCreated '[PlainText] NoContent
-- delivery endpoint of newly published posts of the relay's instance
:<|> "relay" :> "subscribers" :> Capture "senderID" Integer :> ReqBody '[PlainText] Txt.Text :> PostNoContent '[PlainText] Txt.Text
-- endpoint for delivering the subscriptions and outstanding queue
:<|> "post" :> Capture "postid" Txt.Text :> Get '[PlainText] Txt.Text
-- fetch endpoint for posts, full post ID is http://$domain/post/$postid
:<|> "posts" :> ReqBody '[PlainText] Txt.Text :> Post '[PlainText] Txt.Text
-- endpoint for fetching multiple posts at once
:<|> "posts" :> "inbox" :> ReqBody '[PlainText] Txt.Text :> PutCreated '[PlainText] NoContent
-- delivery endpoint of newly published posts of the relay's instance
:<|> "tags" :> Capture "hashtag" Txt.Text :> ReqBody '[PlainText] Txt.Text :> PostCreated '[PlainText] Txt.Text
-- delivery endpoint for posts of $tag at subscribing instance
:<|> "tags" :> Capture "hashtag" Txt.Text :> "subscribe" :> Header "Origin" Txt.Text :> Get '[PlainText] Integer
-- endpoint for subscribing the instance specified in
-- the Origin header to $hashtag.
-- Returns subscription lease time in seconds.
:<|> "tags" :> Capture "hashtag" Txt.Text :> "unsubscribe" :> Header "Origin" Txt.Text :> Get '[PlainText] Txt.Text
-- endpoint for unsubscribing the instance specified in
-- the Origin header to $hashtag
postServer :: DHT d => PostService d -> Server PostServiceAPI
postServer service = relayInbox service
:<|> subscriptionDelivery service
@ -177,27 +181,24 @@ postServer service = relayInbox service
:<|> tagUnsubscribe service
-- | delivery endpoint: receive posts of a handled tag and enqueue them for relaying
relayInbox :: DHT d => PostService d -> Hashtag -> Text -> Handler NoContent
relayInbox :: DHT d => PostService d -> Hashtag -> Txt.Text -> Handler NoContent
relayInbox serv tag posts = do
let
-- skip checking whether the post actually contains the tag, just drop full post
postIDs = head . Txt.splitOn "," <$> Txt.lines posts
-- if tag is not in own responsibility, return a 410 Gone
responsible <- liftIO $ isResponsibleFor (baseDHT serv) (hashtagToId tag)
responsible <- liftIO $ isResponsibleFor (baseDHT serv) (genKeyID . Txt.unpack $ tag)
if responsible
then pure ()
else
throwError $ err410 { errBody = "Relay is not responsible for this tag"}
(throwError $ err410 { errBody = "Relay is not responsible for this tag"})
broadcastChan <- liftIO $ atomically $ getTagBroadcastChannel serv tag
maybe
-- if noone subscribed to the tag, nothing needs to be done
(pure ())
-- otherwise enqueue posts into broadcast queue of the tag
(\queue -> do
(\queue ->
liftIO $ forM_ postIDs (atomically . writeTChan queue)
-- report the received post for statistic purposes
liftIO . atomically . writeTQueue (statsQueue serv) $ StatsEvent RelayReceiveEvent (length postIDs) (hashtagToId tag)
)
broadcastChan
pure NoContent
@ -208,8 +209,7 @@ newtype UnhandledTagException = UnhandledTagException String
instance Exception UnhandledTagException
-- | delivery endpoint: receives a list of subscribers of tags and their outstanding queues for migration
subscriptionDelivery :: DHT d => PostService d -> Integer -> Text -> Handler Text
subscriptionDelivery :: DHT d => PostService d -> Integer -> Txt.Text -> Handler Txt.Text
subscriptionDelivery serv senderID subList = do
let
tagSubs = Txt.lines subList
@ -221,7 +221,7 @@ subscriptionDelivery serv senderID subList = do
-- not-handled tag occurs, this results in a single large transaction.
-- Hopefully the performance isn't too bad.
res <- liftIO . atomically $ (foldM (\_ tag' -> do
responsible <- isResponsibleForSTM (baseDHT serv) (hashtagToId tag')
responsible <- isResponsibleForSTM (baseDHT serv) (genKeyID . Txt.unpack $ tag')
if responsible
then processTag (subscribers serv) tag'
else throwSTM $ UnhandledTagException (Txt.unpack tag' <> " not handled by this relay")
@ -243,7 +243,7 @@ subscriptionDelivery serv senderID subList = do
Right _ -> pure ""
-- TODO: check and only accept tags in own (future?) responsibility
where
processTag :: TVar RelayTags -> Text -> STM ()
processTag :: TVar RelayTags -> Txt.Text -> STM ()
processTag subscriberSTM tagData = do
let
tag:subText:lease:posts:_ = Txt.splitOn "," tagData
@ -254,62 +254,57 @@ subscriptionDelivery serv senderID subList = do
enqueueSubscription subscriberSTM (normaliseTag tag) sub postList leaseTime
-- | endpoint for fetching a post by its ID
postFetch :: PostService d -> Text -> Handler Text
postFetch serv _ = do
-- decision: for saving memory do not store published posts, just
-- pretend there is a post for each requested ID
liftIO . atomically . writeTQueue (statsQueue serv) $ StatsEvent IncomingPostFetchEvent 1 0 -- tag fetched for is irrelevant
pure placeholderPost
postFetch :: PostService d -> Txt.Text -> Handler Txt.Text
postFetch serv postID = do
postSet <- liftIO . readTVarIO . ownPosts $ serv
if HSet.member postID postSet
-- decision: always return the same placeholder post
then pure placeholderPost
else throwError $ err404 { errBody = "No post found with this ID" }
-- | endpoint for fetching multiple posts of this instance by their IDs
postMultiFetch :: PostService d -> Text -> Handler Text
postMultiFetch :: PostService d -> Txt.Text -> Handler Txt.Text
postMultiFetch serv postIDs = do
let
idList = Txt.lines postIDs
-- decision: for saving memory do not store published posts, just
-- pretend there is a post for each requested ID
response = foldl (\response' _ ->
placeholderPost <> "\n" <> response'
let idList = Txt.lines postIDs
postSet <- liftIO . readTVarIO . ownPosts $ serv
-- look up existence of all given post IDs, fail if even one is missing
foldM (\response postID ->
if HSet.member postID postSet
then pure $ placeholderPost <> "\n" <> response
else throwError $ err404 { errBody = "No post found with this ID" }
) "" idList
liftIO . atomically . writeTQueue (statsQueue serv) $ StatsEvent IncomingPostFetchEvent (length idList) 0 -- tag fetched for is irrelevant
pure response
-- | delivery endpoint: inbox for initially publishing a post at an instance
postInbox :: PostService d -> Text -> Handler NoContent
postInbox :: PostService d -> Txt.Text -> Handler NoContent
postInbox serv post = do
-- extract contained hashtags
let
containedTags = fmap (normaliseTag . Txt.tail) . filter ((==) '#' . Txt.head) . Txt.words $ post
-- generate post ID
postId <- liftIO $ Txt.pack . show <$> (randomRIO (0, 2^(128::Integer)-1) :: IO Integer)
-- decision: for saving memory do not store published post IDs, just deliver a post for any requested ID
-- add ID to own posts
liftIO . atomically $ modifyTVar' (ownPosts serv) (HSet.insert postId)
-- enqueue a relay job for each tag
liftIO $ forM_ (containedTags :: [Text]) (\tag ->
liftIO $ forM_ (containedTags :: [Txt.Text]) (\tag ->
atomically $ writeTQueue (relayInQueue serv) (tag, postId, post)
)
pure NoContent
-- | delivery endpoint: receive postIDs of a certain subscribed hashtag
tagDelivery :: PostService d -> Text -> Text -> Handler Text
tagDelivery :: PostService d -> Txt.Text -> Txt.Text -> Handler Txt.Text
tagDelivery serv hashtag posts = do
let postIDs = Txt.lines posts
subscriptions <- liftIO . readTVarIO . ownSubscriptions $ serv
if isJust (HMap.lookup (hashtagToId hashtag) subscriptions)
if isJust (HMap.lookup (genKeyID . Txt.unpack $ hashtag) subscriptions)
then -- TODO: increase a counter/ statistics for received posts of this tag
liftIO $ forM_ postIDs $ atomically . writeTQueue (postFetchQueue serv)
else -- silently drop posts from unsubscribed tags
pure ()
pure $ "Received a postID for tag " <> hashtag
-- | receive subscription requests to a handled hashtag
tagSubscribe :: DHT d => PostService d -> Text -> Maybe Text -> Handler Integer
tagSubscribe :: DHT d => PostService d -> Txt.Text -> Maybe Txt.Text -> Handler Integer
tagSubscribe serv hashtag origin = do
responsible <- liftIO $ isResponsibleFor (baseDHT serv) (hashtagToId hashtag)
responsible <- liftIO $ isResponsibleFor (baseDHT serv) (genKeyID . Txt.unpack $ hashtag)
if not responsible
-- GONE if not responsible
then throwError err410 { errBody = "not responsible for this tag" }
@ -323,14 +318,12 @@ tagSubscribe serv hashtag origin = do
let leaseTime = now + confSubscriptionExpiryTime (serviceConf serv)
-- setup subscription entry
_ <- liftIO . atomically $ setupSubscriberChannel (subscribers serv) hashtag (BSU.toString $ HTTP.host req, HTTP.port req) leaseTime
--liftIO . putStrLn $ "just got a subscription to " <> Txt.unpack hashtag
pure $ round leaseTime
-- | receive and handle unsubscription requests regarding a handled tag
tagUnsubscribe :: DHT d => PostService d -> Text -> Maybe Text -> Handler Text
tagUnsubscribe :: DHT d => PostService d -> Txt.Text -> Maybe Txt.Text -> Handler Txt.Text
tagUnsubscribe serv hashtag origin = do
responsible <- liftIO $ isResponsibleFor (baseDHT serv) (hashtagToId hashtag)
responsible <- liftIO $ isResponsibleFor (baseDHT serv) (genKeyID . Txt.unpack $ hashtag)
if not responsible
-- GONE if not responsible
then throwError err410 { errBody = "not responsible for this tag" }
@ -348,15 +341,8 @@ tagUnsubscribe serv hashtag origin = do
clientAPI :: Proxy PostServiceAPI
clientAPI = Proxy
relayInboxClient
:<|> subscriptionDeliveryClient
:<|> postFetchClient
:<|> postMultiFetchClient
:<|> postInboxClient
:<|> tagDeliveryClient
:<|> tagSubscribeClient
:<|> tagUnsubscribeClient
= client clientAPI
relayInboxClient :<|> subscriptionDeliveryClient :<|> postFetchClient :<|> postMultiFetchClient :<|> postInboxClient :<|> tagDeliveryClient :<|> tagSubscribeClient :<|> tagUnsubscribeClient = client clientAPI
-- | Deliver the subscriber list of all hashtags in the interval [fromTag, toTag]
@ -369,7 +355,7 @@ clientDeliverSubscriptions :: PostService d
-> (String, Int) -- ^ hostname and port of instance to deliver to
-> IO (Either String ()) -- Either signals success or failure
clientDeliverSubscriptions serv fromNode fromKey toKey (toHost, toPort) = do
-- collect tag interval
-- collect tag intearval
intervalTags <- takeRMapSuccessorsFromTo fromKey toKey <$> readTVarIO (subscribers serv)
-- returns a [ (TagSubscribersSTM, TChan PostID, Hashtag) ]
-- extract subscribers and posts
@ -399,7 +385,7 @@ clientDeliverSubscriptions serv fromNode fromKey toKey (toHost, toPort) = do
Right _ -> do
atomically $
modifyTVar' (subscribers serv) $ \tagMap ->
foldr deleteRMapEntry tagMap ((\(_, _, t) -> hashtagToId t) <$> intervalTags)
foldr deleteRMapEntry tagMap ((\(_, _, t) -> genKeyID . Txt.unpack $ t) <$> intervalTags)
pure . Right $ ()
where
channelGetAll :: TChan a -> STM [a]
@ -410,8 +396,7 @@ clientDeliverSubscriptions serv fromNode fromKey toKey (toHost, toPort) = do
maybe (pure acc) (\x -> channelGetAll' chan (x:acc)) haveRead
-- | Subscribe the client to the given hashtag. On success it returns the given lease time,
-- but also records the subscription in its own data structure.
-- | Subscribe the client to the given hashtag. On success it returns the given lease time.
clientSubscribeTo :: DHT d => PostService d -> Hashtag -> IO (Either String Integer)
clientSubscribeTo serv tag = do
lookupRes <- lookupKey (baseDHT serv) (Txt.unpack tag)
@ -426,13 +411,9 @@ clientSubscribeTo serv tag = do
Left (FailureResponse _ fresp)
|(HTTPT.statusCode . responseStatusCode $ fresp) == 410 && allowRetry -> do -- responsibility gone, force new lookup
newRes <- forceLookupKey (baseDHT serv) (Txt.unpack tag)
--putStrLn $ "failed subscribing to " <> Txt.unpack tag <> " on " <> foundHost
doSubscribe newRes False
Left err -> pure . Left . show $ err
Right lease -> do
atomically . modifyTVar' (ownSubscriptions serv) $ HMap.insert (hashtagToId tag) (fromInteger lease)
--putStrLn $ "just subscribed to " <> Txt.unpack tag <> " on " <> foundHost
pure . Right $ lease
Right lease -> pure . Right $ lease
)
lookupResponse
@ -454,9 +435,7 @@ clientUnsubscribeFrom serv tag = do
newRes <- forceLookupKey (baseDHT serv) (Txt.unpack tag)
doUnsubscribe newRes False
Left err -> pure . Left . show $ err
Right _ -> do
atomically . modifyTVar' (ownSubscriptions serv) $ HMap.delete (hashtagToId tag)
pure . Right $ ()
Right _ -> pure . Right $ ()
)
lookupResponse
@ -466,11 +445,11 @@ clientUnsubscribeFrom serv tag = do
-- the post to the responsible relays.
-- As the initial publishing isn't done by a specific relay (but *to* a specific relay
-- instead), the function does *not* take a PostService as argument.
clientPublishPost :: HTTP.Manager -- ^ for better performance, a shared HTTP manager has to be provided
-> String -- ^ hostname
-> Int -- ^ port
-> PostContent -- ^ post content
-> IO (Either String ()) -- ^ error or success
clientPublishPost :: HTTP.Manager -- for better performance, a shared HTTP manager has to be provided
-> String -- hostname
-> Int -- port
-> PostContent -- post content
-> IO (Either String ()) -- error or success
clientPublishPost httpman hostname port postC = do
resp <- runClientM (postInboxClient postC) (mkClientEnv httpman (BaseUrl Http hostname port ""))
pure . bimap show (const ()) $ resp
@ -513,7 +492,7 @@ setupSubscriberChannel tagMapSTM tag subscriber leaseTime = do
broadcastChan <- newBroadcastTChan
tagOutChan <- dupTChan broadcastChan
newSubMapSTM <- newTVar $ HMap.singleton subscriber (tagOutChan, leaseTime)
writeTVar tagMapSTM $ addRMapEntry (hashtagToId tag) (newSubMapSTM, broadcastChan, tag) tagMap
writeTVar tagMapSTM $ addRMapEntry (genKeyID . Txt.unpack $ tag) (newSubMapSTM, broadcastChan, tag) tagMap
pure tagOutChan
Just (foundSubMapSTM, broadcastChan, _) -> do
-- otherwise use the existing subscriber map
@ -541,7 +520,7 @@ deleteSubscription tagMapSTM tag subscriber = do
-- if there are no subscriptions for the tag anymore, remove its
-- data sttructure altogether
if HMap.null newSubMap
then writeTVar tagMapSTM $ deleteRMapEntry (hashtagToId tag) tagMap
then writeTVar tagMapSTM $ deleteRMapEntry (genKeyID . Txt.unpack $ tag) tagMap
-- otherwise just remove the subscription of that node
else writeTVar foundSubMapSTM newSubMap
@ -562,39 +541,12 @@ getTagBroadcastChannel serv tag = do
-- | look up the subscription data of a tag
lookupTagSubscriptions :: Hashtag -> RingMap NodeID a -> Maybe a
lookupTagSubscriptions tag = rMapLookup (hashtagToId tag)
lookupTagSubscriptions tag = rMapLookup (genKeyID . Txt.unpack $ tag)
-- normalise the unicode representation of a string to NFC and convert to lower case
normaliseTag :: Text -> Text
normaliseTag = Txt.toLower . Txt.fromStrict . normalize NFC . Txt.toStrict
-- | convert a hashtag to its representation on the DHT
hashtagToId :: Hashtag -> NodeID
hashtagToId = genKeyID . Txt.unpack
readUpToTChan :: Int -> TChan a -> STM [a]
readUpToTChan 0 _ = pure []
readUpToTChan n chan = do
readFromChan <- tryReadTChan chan
case readFromChan of
Nothing -> pure []
Just val -> do
moreReads <- readUpToTChan (pred n) chan
pure (val:moreReads)
readUpToTQueue :: Int -> TQueue a -> STM [a]
readUpToTQueue 0 _ = pure []
readUpToTQueue n q = do
readFromQueue <- tryReadTQueue q
case readFromQueue of
Nothing -> pure []
Just val -> do
moreReads <- readUpToTQueue (pred n) q
pure (val:moreReads)
-- normalise the unicode representation of a string to NFC
normaliseTag :: Txt.Text -> Txt.Text
normaliseTag = Txt.fromStrict . normalize NFC . Txt.toStrict
-- | define how to convert all showable types to PlainText
-- No idea what I'm doing with these overlappable instances though ¯\_(ツ)_/¯
@ -607,78 +559,28 @@ instance {-# OVERLAPPABLE #-} Read a => MimeUnrender PlainText a where
-- ====== worker threads ======
-- TODO: make configurable
numParallelDeliveries = 10
launchWorkerThreads :: DHT d => PostService d -> IO ()
launchWorkerThreads serv = concurrently_
(processIncomingPosts serv)
$ concurrently_
(purgeSubscriptionsThread serv)
$ concurrently_
(fetchTagPosts serv)
(relayWorker serv)
-- | periodically remove expired subscription entries from relay subscribers
purgeSubscriptionsThread :: PostService d -> IO ()
purgeSubscriptionsThread serv = forever $ do
-- read config
now <- getPOSIXTime
let
purgeInterval = confSubscriptionExpiryTime (serviceConf serv) / 10
-- no need to atomically lock this, as newly incoming subscriptions do not
-- need to be purged
tagMap <- readTVarIO $ subscribers serv
forM_ tagMap $ \(subscriberMapSTM, _, _) ->
-- but each subscriberMap needs to be modified atomically
atomically . modifyTVar' subscriberMapSTM $ HMap.filter (\(_, ts) -> ts > now)
threadDelay $ fromEnum purgeInterval `div` 10^6
-- | process the pending relay inbox of incoming posts from the internal queue:
-- Look up responsible relay node for given hashtag and forward post to it
processIncomingPosts :: DHT d => PostService d -> IO ()
processIncomingPosts serv = forever $ do
-- blocks until available
deliveriesToProcess <- atomically $ do
readResult <- readUpToTQueue numParallelDeliveries $ relayInQueue serv
if null readResult
then retry
else pure readResult
runningJobs <- forM deliveriesToProcess $ \(tag, pID, pContent) -> async $ do
let pIdUri = "http://" <> (Txt.pack . confServiceHost . serviceConf $ serv) <> ":" <> (fromString . show . confServicePort . serviceConf $ serv) <> "/post/" <> pID
lookupRes <- lookupKey (baseDHT serv) (Txt.unpack tag)
case lookupRes of
-- no vserver active => wait and retry
Nothing -> threadDelay (10 * 10^6) >> pure (Left "no vserver active")
Just (responsibleHost, responsiblePort) -> do
resp <- runClientM (relayInboxClient tag $ pIdUri <> "," <> pContent) (mkClientEnv (httpMan serv) (BaseUrl Http responsibleHost (fromIntegral responsiblePort) ""))
case resp of
Left err -> do
-- 410 error indicates outdated responsibility mapping
-- Simplification: just invalidate the mapping entry on all errors, force a re-lookup and re-queue the post
-- TODO: keep track of maximum retries
_ <- forceLookupKey (baseDHT serv) (Txt.unpack tag)
atomically . writeTQueue (relayInQueue serv) $ (tag, pID, pContent)
pure . Left $ "Error: " <> show err
Right _ -> do
-- idea for the experiment: each post publication makes the initial posting instance subscribe to all contained tags
now <- getPOSIXTime
subscriptionStatus <- HMap.lookup (hashtagToId tag) <$> readTVarIO (ownSubscriptions serv)
-- if not yet subscribed or subscription expires within 5 minutes, (re)subscribe to tag
when (maybe True (\subLease -> now - subLease < 300) subscriptionStatus) $
void $ clientSubscribeTo serv tag
-- for evaluation, return the tag of the successfully forwarded post
pure $ Right tag
-- collect async results
results <- mapM waitCatch runningJobs
-- report the count of published posts for statistics
atomically . writeTQueue (statsQueue serv) $ StatsEvent PostPublishEvent (length . rights $ results) 0 -- hashtag published to doesn't matter
pure ()
-- TODO: process multiple in parallel
(tag, pID, pContent) <- atomically . readTQueue $ relayInQueue serv
lookupRes <- lookupKey (baseDHT serv) (Txt.unpack tag)
case lookupRes of
-- no vserver active => wait and retry
Nothing -> threadDelay $ 10 * 10^6
Just (responsibleHost, responsiblePort) -> do
resp <- runClientM (relayInboxClient tag $ pID <> "," <> pContent) (mkClientEnv (httpMan serv) (BaseUrl Http responsibleHost (fromIntegral responsiblePort) ""))
case resp of
Left err -> do
putStrLn $ "Error: " <> show err
-- 410 error indicates outdated responsibility mapping
-- Simplification: just invalidate the mapping entry on all errors, force a re-lookup and re-queue the post
-- TODO: keep track of maximum retries
_ <- forceLookupKey (baseDHT serv) (Txt.unpack tag)
atomically . writeTQueue (relayInQueue serv) $ (tag, pID, pContent)
Right yay -> putStrLn $ "Yay! " <> show yay
-- | process the pending fetch jobs of delivered post IDs: Delivered posts are tried to be fetched from their URI-ID
@ -688,197 +590,18 @@ fetchTagPosts serv = forever $ do
-- TODO: batching, retry
-- TODO: process multiple in parallel
pIdUri <- atomically . readTQueue $ postFetchQueue serv
fetchReq <- HTTP.parseRequest . Txt.unpack $ pIdUri
fetchReq <- HTTP.parseRequest . Txt.unpack $pIdUri
resp <- try $ HTTP.httpLbs fetchReq (httpMan serv) :: IO (Either HTTP.HttpException (HTTP.Response BSUL.ByteString))
case resp of
Right response ->
-- TODO error handling, retry
--if HTTPT.statusCode (HTTP.responseStatus response) == 200
-- then
-- -- success, TODO: statistics
-- else
pure ()
if HTTPT.statusCode (HTTP.responseStatus response) == 200
then
-- success, TODO: statistics
putStrLn "post fetch success"
else
-- TODO error handling, retry
pure ()
Left _ ->
-- TODO error handling, retry
pure ()
relayWorker :: PostService d -> IO ()
relayWorker serv = forever $ do
-- atomically (to be able to retry) fold a list of due delivery actions
jobsToProcess <- atomically $ do
subscriptionMap <- readTVar $ subscribers serv
jobList <- D.toList <$> foldM (\jobAcc (subscriberMapSTM, _, tag) -> do
subscriberMap <- readTVar subscriberMapSTM
foldM (\jobAcc' ((subHost, subPort), (postChan, _)) -> do
postsToDeliver <- readUpToTChan 500 postChan
let postDeliveryAction = runClientM (tagDeliveryClient tag (Txt.unlines postsToDeliver)) (mkClientEnv (httpMan serv) (BaseUrl Http subHost (fromIntegral subPort) ""))
-- append relay push job to job list
pure $ if not (null postsToDeliver)
then jobAcc' `D.snoc` (do
deliveryResult <- postDeliveryAction
either
(const $ pure ())
-- on successful push, record that event for statistics
(const . atomically . writeTQueue (statsQueue serv) $ StatsEvent RelayDeliveryEvent (length postsToDeliver) (hashtagToId tag))
deliveryResult
pure deliveryResult
)
else jobAcc'
) jobAcc $ HMap.toList subscriberMap
) D.empty subscriptionMap
-- if no relay jobs, then retry
if null jobList
then retry
else pure jobList
-- when processing the list, send several deliveries in parallel
forM_ (chunksOf numParallelDeliveries jobsToProcess) $ \jobset -> do
runningJobs <- mapM async jobset
-- so far just dropping failed attempts, TODO: retry mechanism
results <- mapM waitCatch runningJobs
let
successfulResults = rights results
unsuccessfulResults = lefts results
unless (null unsuccessfulResults) $ putStrLn ("ERR: " <> show (length unsuccessfulResults) <> " failed deliveries!")
putStrLn $ "successfully relayed " <> show (length successfulResults)
pure ()
-- ======= statistics/measurement and logging =======
data StatsEventType = PostPublishEvent
| RelayReceiveEvent
| RelayDeliveryEvent
| IncomingPostFetchEvent
deriving (Enum, Show, Eq)
-- | Represents measurement event of a 'StatsEventType' with a count relevant for a certain key
data StatsEvent = StatsEvent StatsEventType Int NodeID
deriving (Show, Eq)
-- | measured rates of relay performance
-- TODO: maybe include other metrics in here as well, like number of subscribers?
data RelayStats = RelayStats
{ relayReceiveRates :: RingMap NodeID Double
-- ^ rate of incoming posts in the responsibility of this relay
, relayDeliveryRates :: RingMap NodeID Double
-- ^ rate of relayed outgoing posts
, postFetchRate :: Double -- no need to differentiate between tags
-- ^ number of post-fetches delivered
, postPublishRate :: Double
-- ^ rate of initially publishing posts through this instance
}
deriving (Show, Eq)
launchStatsThreads :: PostService d -> IO ()
launchStatsThreads serv = do
-- create shared accumulator
sharedAccum <- newTVarIO emptyStats
concurrently_
(accumulateStatsThread sharedAccum $ statsQueue serv)
(evaluateStatsThread serv sharedAccum)
-- | Read stats events from queue and add them to a shared accumulator.
-- Instead of letting the events accumulate in the queue and allocate linear memory, immediately fold the result.
accumulateStatsThread :: TVar RelayStats -> TQueue StatsEvent -> IO ()
accumulateStatsThread statsAccumulator statsQ = forever $ do
-- blocks until stats event arrives
event <- atomically $ readTQueue statsQ
-- add the event number to current accumulator
atomically $ modifyTVar' statsAccumulator $ statsAdder event
-- | add incoming stats events to accumulator value
statsAdder :: StatsEvent -> RelayStats -> RelayStats
statsAdder event stats = case event of
StatsEvent PostPublishEvent num _ ->
stats {postPublishRate = fromIntegral num + postPublishRate stats}
StatsEvent RelayReceiveEvent num key ->
stats {relayReceiveRates = sumIfEntryExists key (fromIntegral num) (relayReceiveRates stats)}
StatsEvent RelayDeliveryEvent num key ->
stats {relayDeliveryRates = sumIfEntryExists key (fromIntegral num) (relayDeliveryRates stats)}
StatsEvent IncomingPostFetchEvent num _ ->
stats {postFetchRate = fromIntegral num + postFetchRate stats}
where
sumIfEntryExists = addRMapEntryWith (\newVal oldVal ->
let toInsert = fromJust $ extractRingEntry newVal
in
case oldVal of
KeyEntry n -> KeyEntry (n + toInsert)
ProxyEntry pointer (Just (KeyEntry n)) -> ProxyEntry pointer (Just (KeyEntry $ n + toInsert))
ProxyEntry pointer Nothing -> ProxyEntry pointer (Just newVal)
_ -> error "RingMap nested too deeply"
)
-- Periodically exchange the accumulated statistics with empty ones, evaluate them
-- and make them the current statistics of the service.
evaluateStatsThread :: PostService d -> TVar RelayStats -> IO ()
evaluateStatsThread serv statsAcc = getPOSIXTime >>= loop
where
loop previousTs = do
threadDelay $ confStatsEvalDelay (serviceConf serv)
-- get and reset the stats accumulator
summedStats <- atomically $ do
stats <- readTVar statsAcc
writeTVar statsAcc emptyStats
pure stats
-- as the transaction might retry several times, current time needs to
-- be read afterwards
now <- getPOSIXTime
-- evaluate stats rate and replace server stats
-- persistently store in a TVar so it can be retrieved later by the DHT
let timePassed = (now - previousTs) * fromIntegral (confSpeedupFactor $ serviceConf serv)
rateStats = evaluateStats timePassed summedStats
atomically $ writeTVar (loadStats serv) rateStats
-- and now what? write a log to file
-- format: total relayReceiveRates;total relayDeliveryRates;postFetchRate;postPublishRate; subscriberSum
-- later: current (reported) load, target load
subscriberSum <- sumSubscribers
TxtI.hPutStrLn (logFileHandle serv) $
format (fixed 9 % ";" % fixed 20 % ";" % fixed 20 % ";" % fixed 20 % ";" % fixed 20 % ";" % int )
(realToFrac now :: Double)
(sum . relayReceiveRates $ rateStats)
(sum . relayDeliveryRates $ rateStats)
(postPublishRate rateStats)
(postFetchRate rateStats)
subscriberSum
loop now
sumSubscribers = do
tagMap <- readTVarIO $ subscribers serv
foldM (\subscriberSum (subscriberMapSTM, _, _) -> do
subscriberMap <- readTVarIO subscriberMapSTM
pure $ subscriberSum + HMap.size subscriberMap
)
0 tagMap
-- | Evaluate the accumulated statistic events: Currently mostly calculates the event
-- rates by dividing through the collection time frame
evaluateStats :: POSIXTime -> RelayStats -> RelayStats
evaluateStats timeInterval summedStats =
-- first sum all event numbers, then divide through number of seconds passed to
-- get rate per second
RelayStats
{ relayReceiveRates = (/ intervalSeconds) <$> relayReceiveRates summedStats
, relayDeliveryRates = (/ intervalSeconds) <$> relayDeliveryRates summedStats
, postPublishRate = postPublishRate summedStats / intervalSeconds
, postFetchRate = postFetchRate summedStats / intervalSeconds
}
where
intervalSeconds = realToFrac timeInterval
emptyStats :: RelayStats
emptyStats = RelayStats
{ relayReceiveRates = emptyRMap
, relayDeliveryRates = emptyRMap
, postFetchRate = 0
, postPublishRate = 0
}

View file

@ -1,37 +0,0 @@
{-# LANGUAGE DataKinds #-}
{-# LANGUAGE FlexibleInstances #-}
{-# LANGUAGE InstanceSigs #-}
{-# LANGUAGE MultiParamTypeClasses #-}
{-# LANGUAGE OverloadedStrings #-}
{-# LANGUAGE RankNTypes #-}
{-# LANGUAGE TypeOperators #-}
module Hash2Pub.PostService.API where
import Data.Text.Lazy (Text)
import Servant
type PostServiceAPI = "relay" :> "inbox" :> Capture "hashtag" Text :> ReqBody '[PlainText] Text :> PutCreated '[PlainText] NoContent
-- delivery endpoint at responsible relay for delivering posts of $tag for distribution
:<|> "relay" :> "subscribers" :> Capture "senderID" Integer :> ReqBody '[PlainText] Text :> PostNoContent '[PlainText] Text
-- endpoint for delivering the subscriptions and outstanding queue
:<|> "post" :> Capture "postid" Text :> Get '[PlainText] Text
-- fetch endpoint for posts, full post ID is http://$domain/post/$postid
:<|> "posts" :> ReqBody '[PlainText] Text :> Post '[PlainText] Text
-- endpoint for fetching multiple posts at once
:<|> "posts" :> "inbox" :> ReqBody '[PlainText] Text :> PutCreated '[PlainText] NoContent
-- delivery endpoint of newly published posts of the relay's instance
:<|> "tags" :> Capture "hashtag" Text :> ReqBody '[PlainText] Text :> PostCreated '[PlainText] Text
-- delivery endpoint for posts of $tag at subscribing instance
:<|> "tags" :> Capture "hashtag" Text :> "subscribe" :> Header "Origin" Text :> Get '[PlainText] Integer
-- endpoint for subscribing the instance specified in
-- the Origin header to $hashtag.
-- Returns subscription lease time in seconds.
:<|> "tags" :> Capture "hashtag" Text :> "unsubscribe" :> Header "Origin" Text :> Get '[PlainText] Text
-- endpoint for unsubscribing the instance specified in
-- the Origin header to $hashtag
-- | needed for guiding type inference
exposedPostServiceAPI :: Proxy PostServiceAPI
exposedPostServiceAPI = Proxy

View file

@ -1,5 +1,7 @@
module Hash2Pub.ProtocolTypes where
import qualified Data.Map as Map
import Data.Maybe (mapMaybe)
import qualified Data.Set as Set
import Data.Time.Clock.POSIX (POSIXTime)

View file

@ -23,30 +23,7 @@ instance (Bounded k, Ord k, Eq a) => Eq (RingMap k a) where
a == b = getRingMap a == getRingMap b
instance (Bounded k, Ord k, Show k, Show a) => Show (RingMap k a) where
show rmap = shows ("RingMap " :: String) (show $ getRingMap rmap)
instance (Bounded k, Ord k) => Functor (RingMap k) where
-- | map a function over all payload values of a 'RingMap'
fmap f = RingMap . Map.map traversingF . getRingMap
where
traversingF (KeyEntry a) = KeyEntry (f a)
traversingF (ProxyEntry pointer (Just entry)) = ProxyEntry pointer (Just $ traversingF entry)
traversingF (ProxyEntry pointer Nothing) = ProxyEntry pointer Nothing
instance (Bounded k, Ord k) => Foldable (RingMap k) where
foldr f initVal = Map.foldr traversingFR initVal . getRingMap
where
traversingFR (KeyEntry a) acc = f a acc
traversingFR (ProxyEntry _ Nothing) acc = acc
traversingFR (ProxyEntry _ (Just entry)) acc = traversingFR entry acc
foldl f initVal = Map.foldl traversingFL initVal . getRingMap
where
traversingFL acc (KeyEntry a) = f acc a
traversingFL acc (ProxyEntry _ Nothing) = acc
traversingFL acc (ProxyEntry _ (Just entry)) = traversingFL acc entry
show rmap = shows "RingMap " (show $ getRingMap rmap)
-- | entry of a 'RingMap' that holds a value and can also
-- wrap around the lookup direction at the edges of the name space.
@ -156,7 +133,7 @@ rMapLookupPred :: (Bounded k, Ord k, Num k)
rMapLookupPred = lookupWrapper Map.lookupLT Map.lookupLE Backwards
addRMapEntryWith :: (Bounded k, Ord k)
=> (RingEntry k a -> RingEntry k a -> RingEntry k a) -- ^ f new_value mold_value
=> (RingEntry k a -> RingEntry k a -> RingEntry k a)
-> k -- ^ key
-> a -- ^ value
-> RingMap k a
@ -230,7 +207,7 @@ takeEntriesUntil_ :: (Integral i, Bounded k, Ord k)
-> Maybe i -- possible number limit
-> [a]
-> [a]
takeEntriesUntil_ _rmap' _getterFunc' _havingReached _previousEntry (Just remaining) takeAcc
takeEntriesUntil_ rmap' getterFunc' havingReached previousEntry (Just remaining) takeAcc
-- length limit reached
| remaining <= 0 = takeAcc
takeEntriesUntil_ rmap' getterFunc' havingReached previousEntry numLimit takeAcc =