If I am using redis to check whether a unique message has been handled historically or is currently being handled (for preventing Pub/Sub duplicate message handling), and I am not expecting to read this kv pair ever again, should I bother deleting the kv pair at the end of the message handler, or just let the LRU eviction eventually delete it? The processing of the message would take far longer than the delay between duplicate messages.
For context, this is some JS pseudocode of what the processing looks like:
// Message comes in
const messageHandler = (message) => {
const duplicate = checkMessageInRedis(message.ID)
if (duplicate) return
registerMessageInRedis(message.ID)
// ... do some stuff to the message
deleteMessageInRedis(message.ID) // DO I BOTHER WITH THIS??? OR LET EVICTION DELETE IT?
}
I guess the question becomes: Is an eviction delete more resource intensive than a DEL transaction? And if so by how much?
Best approach would be to provide a TTL (time to live) when adding the message to Redis.
i.e. registerMessageInRedis(message.ID, ttl)
This will auto-delete the message after ttl expires and will save from an additional network round trip cost for explicit delete.
Related
I have a piece of code that is essentially executing the following with Infinispan in embedded mode, using version 13.0.0 of the -core and -clustered-lock modules:
#Inject
lateinit var lockManager: ClusteredLockManager
private fun getLock(lockName: String): ClusteredLock {
lockManager.defineLock(lockName)
return lockManager.get(lockName)
}
fun createSession(sessionId: String) {
tryLockCounter.increment()
logger.debugf("Trying to start session %s. trying to acquire lock", sessionId)
Future.fromCompletionStage(getLock(sessionId).lock()).map {
acquiredLockCounter.increment()
logger.debugf("Starting session %s. Got lock", sessionId)
}.onFailure {
logger.errorf(it, "Failed to start session %s", sessionId)
}
}
I take this piece of code and deploy it to kubernetes. I then run it in six pods distributed over six nodes in the same region. The code exposes createSession with random Guids through an API. This API is called and creates sessions in chunks of 500, using a k8s service in front of the pods which means the load gets balanced over the pods. I notice that the execution time to acquire a lock grows linearly with the amount of sessions. In the beginning it's around 10ms, when there's about 20_000 sessions it takes about 100ms and the trend continues in a stable fashion.
I then take the same code and run it, but this time with twelve pods on twelve nodes. To my surprise I see that the performance characteristics are almost identical to when I had six pods. I've been digging in to the code but still haven't figured out why this is, I'm wondering if there's a good reason why infinispan here doesn't seem to perform better with more nodes?
For completeness the configuration of the locks are as follows:
val global = GlobalConfigurationBuilder.defaultClusteredBuilder()
global.addModule(ClusteredLockManagerConfigurationBuilder::class.java)
.reliability(Reliability.AVAILABLE)
.numOwner(1)
and looking at the code the clustered locks is using DIST_SYNC which should spread out the load of the cache onto the different nodes.
UPDATE:
The two counters in the code above are simply micrometer counters. It is through them and prometheus that I can see how the lock creation starts to slow down.
It's correctly observed that there's one lock created per session id, this is per design what we'd like. Our use case is that we want to ensure that a session is running in at least one place. Without going to deep into detail this can be achieved by ensuring that we at least have two pods that are trying to acquire the same lock. The Infinispan library is great in that it tells us directly when the lock holder dies without any additional extra chattiness between pods, which means that we have a "cheap" way of ensuring that execution of the session continues when one pod is removed.
After digging deeper into the code I found the following in CacheNotifierImpl in the core library:
private CompletionStage<Void> doNotifyModified(K key, V value, Metadata metadata, V previousValue,
Metadata previousMetadata, boolean pre, InvocationContext ctx, FlagAffectedCommand command) {
if (clusteringDependentLogic.running().commitType(command, ctx, extractSegment(command, key), false).isLocal()
&& (command == null || !command.hasAnyFlag(FlagBitSets.PUT_FOR_STATE_TRANSFER))) {
EventImpl<K, V> e = EventImpl.createEvent(cache.wired(), CACHE_ENTRY_MODIFIED);
boolean isLocalNodePrimaryOwner = isLocalNodePrimaryOwner(key);
Object batchIdentifier = ctx.isInTxScope() ? null : Thread.currentThread();
try {
AggregateCompletionStage<Void> aggregateCompletionStage = null;
for (CacheEntryListenerInvocation<K, V> listener : cacheEntryModifiedListeners) {
// Need a wrapper per invocation since converter could modify the entry in it
configureEvent(listener, e, key, value, metadata, pre, ctx, command, previousValue, previousMetadata);
aggregateCompletionStage = composeStageIfNeeded(aggregateCompletionStage,
listener.invoke(new EventWrapper<>(key, e), isLocalNodePrimaryOwner));
}
The lock library uses a clustered Listener on the entry modified event, and this one uses a filter to only notify when the key for the lock is modified. It seems to me the core library still has to check this condition on every registered listener, which of course becomes a very big list as the number of sessions grow. I suspect this to be the reason and if it is it would be really really awesome if the core library supported a kind of key filter so that it could use a hashmap for these listeners instead of going through a whole list with all listeners.
I believe you are creating a clustered lock per session id. Is this what you need ? what is the acquiredLockCounter? We are about to deprecate the "lock" method in favour of "tryLock" with timeout since the lock method will block forever if the clustered lock is never acquired. Do you ever unlock the clustered lock in another piece of code? If you shared a complete reproducer of the code will be very helpful for us. Thanks!
I have a use case where I'm aggregating until a TTL hits 0 on the key/value in Redis. As far as I can tell in the documentation, retrieval or a background job triggers all expired keys to be deleted immediately.
Is there anyway I can 'halt' that deletion and retrieve the value at the time of expiry? Or something similar to that effect?
My last question contains some context to my use case: Redis - any way to trigger an event when a value is no longer being actively written to?
I believe I found an answer to my question. Every row would have a corresponding row with with the convention expiry:{key}.
uniqueEventHash: [value1, value2, value3] // no expiry
expiry:uniqueEventHash: {no value} // set TTL to 60
Now, whenever a new value for that uniqueEventHash arrives, I do two things. I append it onto the uniqueEventHash row with an append, and then I also subsequently reset the TTL on expiry:uniqueEventHash to 60.
When events for that uniqueEventHash stop arriving, the second expiry:expiry:uniqueEventHash is deleted and a notification is sent out to a subscriber to EXPIRY events. In the message is the key that expired, which in this case is expiry:uniqueEventHash.
I can then do the following:
// pseudo code
onExpiryEvent(message):
[type, key] = eventMessage.split(':');
aggregatedValue = await getByKey(key);
del(key);
I am using the taskqueue API to send multiple emails is small groups with mailgun. My code looks more or less like this:
class CpMsg(ndb.Model):
group = ndb.KeyProperty()
sent = ndb.BooleanProperty()
#Other properties
def send_mail(messages):
"""Sends a request to mailgun's API"""
# Some code
pass
class MailTask(TaskHandler):
def post(self):
p_key = utils.key_from_string(self.request.get('p'))
msgs = CpMsg.query(
CpMsg.group==p_key,
CpMsg.sent==False).fetch(BATCH_SIZE)
if msgs:
send_mail(msgs)
for msg in msgs:
msg.sent = True
ndb.put_multi(msgs)
#Call the task again in COOLDOWN seconds
The code above has been working fine, but according to the docs, the taskqueue API guarantees that a task is delivered at least once, so tasks should be idempotent. Now, most of the time this would be the case with the above code, since it only gets messages that have the 'sent' property equal to False. The problem is that non ancestor ndb queries are only eventually consistent, which means that if the task is executed twice in quick succession the query may return stale results and include the messages that were just sent.
I thought of including an ancestor for the messages, but since the sent emails will be in the thousands I'm worried that may mean having large entity groups, which have a limited write throughput.
Should I use an ancestor to make the queries? Or maybe there is a way to configure mailgun to avoid sending the same email twice? Should I just accept the risk that in some rare cases a few emails may be sent more than once?
One possible approach to avoid the eventual consistency hurdle is to make the query a keys_only one, then iterate through the message keys to get the actual messages by key lookup (strong consistency), check if msg.sent is True and skip sending those messages in such case. Something along these lines:
msg_keys = CpMsg.query(
CpMsg.group==p_key,
CpMsg.sent==False).fetch(BATCH_SIZE, keys_only=True)
if not msg_keys:
return
msgs = ndb.get_multi(msg_keys)
msgs_to_send = []
for msg in msgs:
if not msg.sent:
msgs_to_send.append(msg)
if msgs_to_send:
send_mail(msgs_to_send)
for msg in msgs_to_send:
msg.sent = True
ndb.put_multi(msgs_to_send)
You'd also have to make your post call transactional (with the #ndb.transactional() decorator).
This should address the duplicates caused by the query eventual consistency. However there still is room for duplicates caused by transaction retries due to datastore contention (or any other reason) - as the send_mail() call isn't idempotent. Sending one message at a time (maybe using the task queue) could reduce the chance of that happening. See also GAE/P: Transaction safety with API calls
I need to design a Redis-driven scalable task scheduling system.
Requirements:
Multiple worker processes.
Many tasks, but long periods of idleness are possible.
Reasonable timing precision.
Minimal resource waste when idle.
Should use synchronous Redis API.
Should work for Redis 2.4 (i.e. no features from upcoming 2.6).
Should not use other means of RPC than Redis.
Pseudo-API: schedule_task(timestamp, task_data). Timestamp is in integer seconds.
Basic idea:
Listen for upcoming tasks on list.
Put tasks to buckets per timestamp.
Sleep until the closest timestamp.
If a new task appears with timestamp less than closest one, wake up.
Process all upcoming tasks with timestamp ≤ now, in batches (assuming
that task execution is fast).
Make sure that concurrent worker wouldn't process same tasks. At the same time, make sure that no tasks are lost if we crash while processing them.
So far I can't figure out how to fit this in Redis primitives...
Any clues?
Note that there is a similar old question: Delayed execution / scheduling with Redis? In this new question I introduce more details (most importantly, many workers). So far I was not able to figure out how to apply old answers here — thus, a new question.
Here's another solution that builds on a couple of others [1]. It uses the redis WATCH command to remove the race condition without using lua in redis 2.6.
The basic scheme is:
Use a redis zset for scheduled tasks and redis queues for ready to run tasks.
Have a dispatcher poll the zset and move tasks that are ready to run into the redis queues. You may want more than 1 dispatcher for redundancy but you probably don't need or want many.
Have as many workers as you want which do blocking pops on the redis queues.
I haven't tested it :-)
The foo job creator would do:
def schedule_task(queue, data, delay_secs):
# This calculation for run_at isn't great- it won't deal well with daylight
# savings changes, leap seconds, and other time anomalies. Improvements
# welcome :-)
run_at = time.time() + delay_secs
# If you're using redis-py's Redis class and not StrictRedis, swap run_at &
# the dict.
redis.zadd(SCHEDULED_ZSET_KEY, run_at, {'queue': queue, 'data': data})
schedule_task('foo_queue', foo_data, 60)
The dispatcher(s) would look like:
while working:
redis.watch(SCHEDULED_ZSET_KEY)
min_score = 0
max_score = time.time()
results = redis.zrangebyscore(
SCHEDULED_ZSET_KEY, min_score, max_score, start=0, num=1, withscores=False)
if results is None or len(results) == 0:
redis.unwatch()
sleep(1)
else: # len(results) == 1
redis.multi()
redis.rpush(results[0]['queue'], results[0]['data'])
redis.zrem(SCHEDULED_ZSET_KEY, results[0])
redis.exec()
The foo worker would look like:
while working:
task_data = redis.blpop('foo_queue', POP_TIMEOUT)
if task_data:
foo(task_data)
[1] This solution is based on not_a_golfer's, one at http://www.saltycrane.com/blog/2011/11/unique-python-redis-based-queue-delay/, and the redis docs for transactions.
You didn't specify the language you're using. You have at least 3 alternatives of doing this without writing a single line of code in Python at least.
Celery has an optional redis broker.
http://celeryproject.org/
resque is an extremely popular redis task queue using redis.
https://github.com/defunkt/resque
RQ is a simple and small redis based queue that aims to "take the good stuff from celery and resque" and be much simpler to work with.
http://python-rq.org/
You can at least look at their design if you can't use them.
But to answer your question - what you want can be done with redis. I've actually written more or less that in the past.
EDIT:
As for modeling what you want on redis, this is what I would do:
queuing a task with a timestamp will be done directly by the client - you put the task in a sorted set with the timestamp as the score and the task as the value (see ZADD).
A central dispatcher wakes every N seconds, checks out the first timestamps on this set, and if there are tasks ready for execution, it pushes the task to a "to be executed NOW" list. This can be done with ZREVRANGEBYSCORE on the "waiting" sorted set, getting all items with timestamp<=now, so you get all the ready items at once. pushing is done by RPUSH.
workers use BLPOP on the "to be executed NOW" list, wake when there is something to work on, and do their thing. This is safe since redis is single threaded, and no 2 workers will ever take the same task.
once finished, the workers put the result back in a response queue, which is checked by the dispatcher or another thread. You can add a "pending" bucket to avoid failures or something like that.
so the code will look something like this (this is just pseudo code):
client:
ZADD "new_tasks" <TIMESTAMP> <TASK_INFO>
dispatcher:
while working:
tasks = ZREVRANGEBYSCORE "new_tasks" <NOW> 0 #this will only take tasks with timestamp lower/equal than now
for task in tasks:
#do the delete and queue as a transaction
MULTI
RPUSH "to_be_executed" task
ZREM "new_tasks" task
EXEC
sleep(1)
I didn't add the response queue handling, but it's more or less like the worker:
worker:
while working:
task = BLPOP "to_be_executed" <TIMEOUT>
if task:
response = work_on_task(task)
RPUSH "results" response
EDit: stateless atomic dispatcher :
while working:
MULTI
ZREVRANGE "new_tasks" 0 1
ZREMRANGEBYRANK "new_tasks" 0 1
task = EXEC
#this is the only risky place - you can solve it by using Lua internall in 2.6
SADD "tmp" task
if task.timestamp <= now:
MULTI
RPUSH "to_be_executed" task
SREM "tmp" task
EXEC
else:
MULTI
ZADD "new_tasks" task.timestamp task
SREM "tmp" task
EXEC
sleep(RESOLUTION)
If you're looking for ready solution on Java. Redisson is right for you. It allows to schedule and execute tasks (with cron-expression support) in distributed way on Redisson nodes using familiar ScheduledExecutorService api and based on Redis queue.
Here is an example. First define a task using java.lang.Runnable interface. Each task can access to Redis instance via injected RedissonClient object.
public class RunnableTask implements Runnable {
#RInject
private RedissonClient redissonClient;
#Override
public void run() throws Exception {
RMap<String, Integer> map = redissonClient.getMap("myMap");
Long result = 0;
for (Integer value : map.values()) {
result += value;
}
redissonClient.getTopic("myMapTopic").publish(result);
}
}
Now it's ready to sumbit it into ScheduledExecutorService:
RScheduledExecutorService executorService = redisson.getExecutorService("myExecutor");
ScheduledFuture<?> future = executorService.schedule(new CallableTask(), 10, 20, TimeUnit.MINUTES);
future.get();
// or cancel it
future.cancel(true);
Examples with cron expressions:
executorService.schedule(new RunnableTask(), CronSchedule.of("10 0/5 * * * ?"));
executorService.schedule(new RunnableTask(), CronSchedule.dailyAtHourAndMinute(10, 5));
executorService.schedule(new RunnableTask(), CronSchedule.weeklyOnDayAndHourAndMinute(12, 4, Calendar.MONDAY, Calendar.FRIDAY));
All tasks are executed on Redisson node.
A combined approach seems plausible:
No new task timestamp may be less than current time (clamp if less). Assuming reliable NTP synch.
All tasks go to bucket-lists at keys, suffixed with task timestamp.
Additionally, all task timestamps go to a dedicated zset (key and score — timestamp itself).
New tasks are accepted from clients via separate Redis list.
Loop: Fetch oldest N expired timestamps via zrangebyscore ... limit.
BLPOP with timeout on new tasks list and lists for fetched timestamps.
If got an old task, process it. If new — add to bucket and zset.
Check if processed buckets are empty. If so — delete list and entrt from zset. Probably do not check very recently expired buckets, to safeguard against time synchronization issues. End loop.
Critique? Comments? Alternatives?
Lua
I made something similar to what's been suggested here, but optimized the sleep duration to be more precise. This solution is good if you have few inserts into the delayed task queue. Here's how I did it with a Lua script:
local laterChannel = KEYS[1]
local nowChannel = KEYS[2]
local currentTime = tonumber(KEYS[3])
local first = redis.call("zrange", laterChannel, 0, 0, "WITHSCORES")
if (#first ~= 2)
then
return "2147483647"
end
local execTime = tonumber(first[2])
local event = first[1]
if (currentTime >= execTime)
then
redis.call("zrem", laterChannel, event)
redis.call("rpush", nowChannel, event)
return "0"
else
return tostring(execTime - currentTime)
end
It uses two "channels". laterChannel is a ZSET and nowChannel is a LIST. Whenever it's time to execute a task, the event is moved from the the ZSET to the LIST. The Lua script with respond with how many MS the dispatcher should sleep until the next poll. If the ZSET is empty, sleep forever. If it's time to execute something, do not sleep(i e poll again immediately). Otherwise, sleep until it's time to execute the next task.
So what if something is added while the dispatcher is sleeping?
This solution works in conjunction with key space events. You basically need to subscribe to the key of laterChannel and whenever there is an add event, you wake up all the dispatcher so they can poll again.
Then you have another dispatcher that uses the blocking left pop on nowChannel. This means:
You can have the dispatcher across multiple instances(i e it's scaling)
The polling is atomic so you won't have any race conditions or double events
The task is executed by any of the instances that are free
There are ways to optimize this even more. For example, instead of returning "0", you fetch the next item from the zset and return the correct amount of time to sleep directly.
Expiration
If you can not use Lua scripts, you can use key space events on expired documents.
Subscribe to the channel and receive the event when Redis evicts it. Then, grab a lock. The first instance to do so will move it to a list(the "execute now" channel). Then you don't have to worry about sleeps and polling. Redis will tell you when it's time to execute something.
execute_later(timestamp, eventId, event) {
SET eventId event EXP timestamp
SET "lock:" + eventId, ""
}
subscribeToEvictions(eventId) {
var deletedCount = DEL eventId
if (deletedCount == 1) {
// move to list
}
}
This however has it own downsides. For example, if you have many nodes, all of them will receive the event and try to get the lock. But I still think it's overall less requests any anything suggested here.
I have a WCF service that posts messages to a private, non-transactional MSMQ queue. I have another WCF service (multi-threaded) that processes the MSMQ messages and inserts them in the database.
My issue is with sequencing. I want the messages to be in certain order. For example MSG-A need to go to the database before MSG-B is inserted. So my current solution for that is very crude and expensive from database perspective.
I am reading the message, if its MSG-B and there is no MSG-A in the database, I throw it back on the message queue and I keep doing that till MSG-A is inserted in the database. But this is a very expensive operation as it involves table scan (SELECT stmt).
The messages are always posted to the queue in sequence.
Short of making my WCF Queue Processing service Single threaded (By setting the service behavior attribute InstanceContextMode to Single), can someone suggest a better solution?
Thanks
Dan
Instead of immediately pushing messages to the DB after taking them out of the queue, keep a list of pending messages in memory. When you get an A or B, check to see if the matching one is in the list. If so, submit them both (in the right order) to the database, and remove the matching one from the list. Otherwise, just add the new message to that list.
If checking for a match is too expensive a task to serialize - I assume you are multithreading for a reason - the you could have another thread process the list. The existing multiple threads read, immediately submit most messages to the DB, but put the As and Bs aside in the (threadsafe) list. The background thread scavenges through that list finding matching As and Bs and when it finds them it submits them in the right order (and removes them from the list).
The bottom line is - since your removing items from the queue with multiple threads, you're going to have to serialize somewhere, in order to ensure ordering. The trick is to minimize the number of times and length of time you spend locked up in serial code.
There might also be something you could do at the database level, with triggers or something, to reorder the entries when it detects this situation. I'm afraid I don't know enough about DB programming to help there.
UPDATE: Assuming the messages contain some id that lets you associate a message 'A' with the correct associated message 'B', the following code will make sure A goes in the database before B. Note that it does not make sure they are adjacent records in the database - there could be other messages between A and B. Also, if for some reason you get an A or B without ever receiving the matching message of the other type, this code will leak memory since it hangs onto the unmatched message forever.
(You could extract those two 'lock'ed blocks into a single subroutine, but I'm leaving it like this for clarity with respect to A and B.)
static private object dictionaryLock = new object();
static private Dictionary<int, MyMessage> receivedA =
new Dictionary<int, MyMessage>();
static private Dictionary<int, MyMessage> receivedB =
new Dictionary<int, MyMessage>();
public void MessageHandler(MyMessage message)
{
MyMessage matchingMessage = null;
if (IsA(message))
{
InsertIntoDB(message);
lock (dictionaryLock)
{
if (receivedB.TryGetValue(message.id, out matchingMessage))
{
receivedB.Remove(message.id);
}
else
{
receivedA.Add(message.id, message);
}
}
if (matchingMessage != null)
{
InsertIntoDB(matchingMessage);
}
}
else if (IsB(message))
{
lock (dictionaryLock)
{
if (receivedA.TryGetValue(message.id, out matchingMessage))
{
receivedA.Remove(message.id);
}
else
{
receivedB.Add(message.id, message);
}
}
if (matchingMessage != null)
{
InsertIntoDB(message);
}
}
else
{
// not A or B, do whatever
}
}
If you're the only client of those queues, you could very easy add a timestamp as a message header (see IDesign sample) and save the Sent On field (kinda like an outlook message) in the database as well. You could process them in the order they were sent (basically you move the sorting logic at the time of consumption).
Hope this helps,
Adrian