You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@kvrocks.apache.org by "webfrank (via GitHub)" <gi...@apache.org> on 2025/02/03 14:44:01 UTC
[I] memtable, bg_error: Corruption: WriteBatch has wrong count [kvrocks]
webfrank opened a new issue, #2766:
URL: https://github.com/apache/kvrocks/issues/2766
### Search before asking
- [x] I had searched in the [issues](https://github.com/apache/kvrocks/issues) and found no similar issues.
### Version
I'm trying to move from Redis to KVRocks as I need a durable storage for a Job Queue system based on python rq.
Everything is working but whenever a restart KVRocks or the python code I receive this error:
E20250203 14:38:03.244154 139655422928576 event_listener.cc:166] [event_listener/background_error] reason: memtable, bg_error: Corruption: WriteBatch has wrong count
I need to start from scratch to solve the issue but this way I loose the data.
### Minimal reproduce step
Start a standalone server and connect a python client doing some continuous operations
### What did you expect to see?
No errors
### What did you see instead?
E20250203 14:38:03.244154 139655422928576 event_listener.cc:166] [event_listener/background_error] reason: memtable, bg_error: Corruption: WriteBatch has wrong count
### Anything Else?
_No response_
### Are you willing to submit a PR?
- [ ] I'm willing to submit a PR!
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@kvrocks.apache.org.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] memtable, bg_error: Corruption: WriteBatch has wrong count [kvrocks]
Posted by "webfrank (via GitHub)" <gi...@apache.org>.
webfrank commented on issue #2766:
URL: https://github.com/apache/kvrocks/issues/2766#issuecomment-2631454545
Hi, I'm using latest 2.11 inside docker.
Another occurrence:
I20250203 16:14:00.264371 140022818768576 compaction_checker.cc:37] [compaction checker] Compact the column family: propagate finished, result: Corruption: WriteBatch has wrong count
Whenever I restart the client
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@kvrocks.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] memtable, bg_error: Corruption: WriteBatch has wrong count [kvrocks]
Posted by "webfrank (via GitHub)" <gi...@apache.org>.
webfrank commented on issue #2766:
URL: https://github.com/apache/kvrocks/issues/2766#issuecomment-2633558088
Hi, yes, same issue on 2.11.0 and 2.10.1. Now, after 4 hours of running the error appears:
I20250204 09:20:33.271770 139658734335680 event_listener.cc:122] [event_listener/flush_begin] column family: default, thread_id: 10, job_id: 4, reason: Write Buffer Full
I20250204 09:20:36.592097 139658734335680 event_listener.cc:181] [event_listener/table_file_created] column family: default, file path: /var/lib/kvrocks/db/000023.sst, file size: 56482559, job_id: 4, reason: flush, status: OK
I20250204 09:20:36.633254 139658734335680 event_listener.cc:129] [event_listener/flush_completed] column family: default, thread_id: 10, job_id: 4, file: /var/lib/kvrocks/db/000023.sst, reason: 6, is_write_slowdown: no, is_write_stall: no, largest seqno: 2010631, smallest seqno: 744991
I20250204 10:14:04.949374 139658734335680 event_listener.cc:122] [event_listener/flush_begin] column family: default, thread_id: 10, job_id: 5, reason: Write Buffer Full
I20250204 10:14:07.789879 139658734335680 event_listener.cc:181] [event_listener/table_file_created] column family: default, file path: /var/lib/kvrocks/db/000027.sst, file size: 39246905, job_id: 5, reason: flush, status: OK
I20250204 10:14:07.831367 139658734335680 event_listener.cc:129] [event_listener/flush_completed] column family: default, thread_id: 10, job_id: 5, file: /var/lib/kvrocks/db/000027.sst, reason: 6, is_write_slowdown: no, is_write_stall: no, largest seqno: 3520980, smallest seqno: 2010632
E20250204 10:51:20.839834 139658688194240 event_listener.cc:166] [event_listener/background_error] reason: memtable, bg_error: Corruption: WriteBatch has wrong count
The workers, after processing data, restarts and reconnect but once this error arise it not possible anymore, I need to wipe db folder.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@kvrocks.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] memtable, bg_error: Corruption: WriteBatch has wrong count [kvrocks]
Posted by "webfrank (via GitHub)" <gi...@apache.org>.
webfrank commented on issue #2766:
URL: https://github.com/apache/kvrocks/issues/2766#issuecomment-2662715075
New case:
```
E20250217 10:35:04.410730 128177347233472 event_listener.cc:166] [event_listener/background_error] reason: memtable, bg_error: Corruption: unknown WriteBatch tag: 101
```
Python traceback
```
Traceback (most recent call last):
File "/usr/local/Caskroom/miniconda/base/envs/meraki/lib/python3.10/site-packages/rq/command.py", line 35, in send_command
connection.publish(PUBSUB_CHANNEL_TEMPLATE % worker_name, json.dumps(payload))
File "/usr/local/Caskroom/miniconda/base/envs/meraki/lib/python3.10/site-packages/redis/commands/core.py", line 5556, in publish
return self.execute_command("PUBLISH", channel, message, **kwargs)
File "/usr/local/Caskroom/miniconda/base/envs/meraki/lib/python3.10/site-packages/redis/client.py", line 548, in execute_command
return conn.retry.call_with_retry(
File "/usr/local/Caskroom/miniconda/base/envs/meraki/lib/python3.10/site-packages/redis/retry.py", line 62, in call_with_retry
return do()
File "/usr/local/Caskroom/miniconda/base/envs/meraki/lib/python3.10/site-packages/redis/client.py", line 549, in <lambda>
lambda: self._send_command_parse_response(
File "/usr/local/Caskroom/miniconda/base/envs/meraki/lib/python3.10/site-packages/redis/client.py", line 525, in _send_command_parse_response
return self.parse_response(conn, command_name, **options)
File "/usr/local/Caskroom/miniconda/base/envs/meraki/lib/python3.10/site-packages/redis/client.py", line 565, in parse_response
response = connection.read_response()
File "/usr/local/Caskroom/miniconda/base/envs/meraki/lib/python3.10/site-packages/redis/connection.py", line 536, in read_response
raise response
redis.exceptions.ResponseError: Corruption: unknown WriteBatch tag: 101
```
This time issuing a PUBLISH command.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@kvrocks.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] memtable, bg_error: Corruption: WriteBatch has wrong count [kvrocks]
Posted by "webfrank (via GitHub)" <gi...@apache.org>.
webfrank commented on issue #2766:
URL: https://github.com/apache/kvrocks/issues/2766#issuecomment-2639050909
Started from scratch
docker-compose.yml
```yaml
services:
kvrocks:
image: apache/kvrocks:2.11.0
command:
- --rocksdb.block_cache_size
- "64"
- --rocksdb.write_buffer_size
- "16"
- --rocksdb.max_write_buffer_number
- "4"
deploy:
resources:
limits:
memory: 512M
volumes:
- kvrocks:/var/lib/kvrocks
networks:
- internal
```
logs
```
I20250206 07:30:56.051099 139693424336128 main.cc:144] kvrocks version 2.11.0 (commit 172d31af)
I20250206 07:30:56.214070 139693424336128 storage.cc:395] [storage] Success to load the data from disk: 48 ms
I20250206 07:30:56.245163 139693424336128 worker.cc:75] [worker] Listening on: 0.0.0.0:6666
I20250206 07:30:56.249971 139693424336128 worker.cc:75] [worker] Listening on: 0.0.0.0:6666
I20250206 07:30:56.250308 139693424336128 worker.cc:75] [worker] Listening on: 0.0.0.0:6666
I20250206 07:30:56.250646 139693424336128 worker.cc:75] [worker] Listening on: 0.0.0.0:6666
I20250206 07:30:56.251055 139693424336128 worker.cc:75] [worker] Listening on: 0.0.0.0:6666
I20250206 07:30:56.251393 139693424336128 worker.cc:75] [worker] Listening on: 0.0.0.0:6666
I20250206 07:30:56.251703 139693424336128 worker.cc:75] [worker] Listening on: 0.0.0.0:6666
I20250206 07:30:56.252146 139693424336128 worker.cc:75] [worker] Listening on: 0.0.0.0:6666
I20250206 07:30:56.252879 139693424336128 worker.cc:586] [worker] Thread #139693272295104 started
I20250206 07:30:56.253008 139693424336128 worker.cc:586] [worker] Thread #139693280687808 started
I20250206 07:30:56.253077 139693424336128 worker.cc:586] [worker] Thread #139693289080512 started
I20250206 07:30:56.253216 139693424336128 worker.cc:586] [worker] Thread #139693343635136 started
I20250206 07:30:56.253332 139693424336128 worker.cc:586] [worker] Thread #139693335242432 started
I20250206 07:30:56.253451 139693424336128 worker.cc:586] [worker] Thread #139693326849728 started
I20250206 07:30:56.253562 139693424336128 worker.cc:586] [worker] Thread #139693318457024 started
I20250206 07:30:56.253669 139693424336128 worker.cc:586] [worker] Thread #139693310064320 started
I20250206 07:30:56.254084 139693424336128 server.cc:240] [server] Ready to accept connections
I20250206 07:31:56.327786 139693255509696 compaction_checker.cc:34] [compaction checker] Start the compact the column family: pubsub
I20250206 07:31:56.327927 139693255509696 compaction_checker.cc:37] [compaction checker] Compact the column family: pubsub finished, result: OK
I20250206 07:31:56.327940 139693255509696 compaction_checker.cc:34] [compaction checker] Start the compact the column family: propagate
I20250206 07:31:56.327975 139693255509696 compaction_checker.cc:37] [compaction checker] Compact the column family: propagate finished, result: OK
E20250206 07:32:01.985579 139693326849728 event_listener.cc:166] [event_listener/background_error] reason: memtable, bg_error: Corruption: WriteBatch has wrong count
```
The error arise once the application connects and issue an HSET command to configure JOB queue.
```
Traceback (most recent call last):
File "/app/main.py", line 34, in <module>
main(redis)
File "/app/main.py", line 25, in main
w.work(max_jobs=1000)
File "/usr/local/lib/python3.10/site-packages/rq/worker.py", line 592, in work
self.bootstrap(logging_level, date_format, log_format)
File "/usr/local/lib/python3.10/site-packages/rq/worker.py", line 927, in bootstrap
self.set_state(WorkerStatus.STARTED)
File "/usr/local/lib/python3.10/site-packages/rq/worker.py", line 806, in set_state
connection.hset(self.key, 'state', state)
File "/usr/local/lib/python3.10/site-packages/redis/commands/core.py", line 5045, in hset
return self.execute_command("HSET", name, *pieces)
File "/usr/local/lib/python3.10/site-packages/redis/client.py", line 559, in execute_command
return self._execute_command(*args, **options)
File "/usr/local/lib/python3.10/site-packages/redis/client.py", line 567, in _execute_command
return conn.retry.call_with_retry(
File "/usr/local/lib/python3.10/site-packages/redis/retry.py", line 62, in call_with_retry
return do()
File "/usr/local/lib/python3.10/site-packages/redis/client.py", line 568, in <lambda>
lambda: self._send_command_parse_response(
File "/usr/local/lib/python3.10/site-packages/redis/client.py", line 542, in _send_command_parse_response
return self.parse_response(conn, command_name, **options)
File "/usr/local/lib/python3.10/site-packages/redis/client.py", line 584, in parse_response
response = connection.read_response()
File "/usr/local/lib/python3.10/site-packages/redis/connection.py", line 616, in read_response
raise response
redis.exceptions.ResponseError: Corruption: WriteBatch has wrong count
```
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@kvrocks.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] memtable, bg_error: Corruption: WriteBatch has wrong count [kvrocks]
Posted by "git-hulk (via GitHub)" <gi...@apache.org>.
git-hulk commented on issue #2766:
URL: https://github.com/apache/kvrocks/issues/2766#issuecomment-2642572657
@webfrank It would be helpful if you could provide a reproduced sample code for this.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@kvrocks.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] memtable, bg_error: Corruption: WriteBatch has wrong count [kvrocks]
Posted by "webfrank (via GitHub)" <gi...@apache.org>.
webfrank commented on issue #2766:
URL: https://github.com/apache/kvrocks/issues/2766#issuecomment-2633071006
Hi, I cannot disclose the code but I will try to extract the logic. What I'm doing is to have KVRocks in a container, one separate container for the job queue controller (the one which schedules jobs on RQ) and 6 workers nodes consuming jobs and updating KVRocks.
What happens is whenever I restart one of them, kvrocks included, connection cannot be stablished and the error arise. I need to restart from scratch every time.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@kvrocks.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] memtable, bg_error: Corruption: WriteBatch has wrong count [kvrocks]
Posted by "PragmaTwice (via GitHub)" <gi...@apache.org>.
PragmaTwice commented on issue #2766:
URL: https://github.com/apache/kvrocks/issues/2766#issuecomment-2632782564
Would you mind to provide detailed steps to reproduce such issue?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@kvrocks.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] memtable, bg_error: Corruption: WriteBatch has wrong count [kvrocks]
Posted by "git-hulk (via GitHub)" <gi...@apache.org>.
git-hulk closed issue #2766: memtable, bg_error: Corruption: WriteBatch has wrong count
URL: https://github.com/apache/kvrocks/issues/2766
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@kvrocks.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] memtable, bg_error: Corruption: WriteBatch has wrong count [kvrocks]
Posted by "git-hulk (via GitHub)" <gi...@apache.org>.
git-hulk closed issue #2766: memtable, bg_error: Corruption: WriteBatch has wrong count
URL: https://github.com/apache/kvrocks/issues/2766
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@kvrocks.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] memtable, bg_error: Corruption: WriteBatch has wrong count [kvrocks]
Posted by "git-hulk (via GitHub)" <gi...@apache.org>.
git-hulk commented on issue #2766:
URL: https://github.com/apache/kvrocks/issues/2766#issuecomment-2662756317
@webfrank Thanks for your report. I'm wondering if it's related to the MULTI-EXEC mode because the main difference is `rq` will depend on that. But I can't reproduce it on my side.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@kvrocks.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] memtable, bg_error: Corruption: WriteBatch has wrong count [kvrocks]
Posted by "git-hulk (via GitHub)" <gi...@apache.org>.
git-hulk commented on issue #2766:
URL: https://github.com/apache/kvrocks/issues/2766#issuecomment-2837849910
@webfrank We finally reproduce this issue in other case: https://github.com/apache/kvrocks/issues/2900#issuecomment-2837848286
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@kvrocks.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] memtable, bg_error: Corruption: WriteBatch has wrong count [kvrocks]
Posted by "git-hulk (via GitHub)" <gi...@apache.org>.
git-hulk commented on issue #2766:
URL: https://github.com/apache/kvrocks/issues/2766#issuecomment-2645031692
@webfrank Thanks for your input. I suppose this bug might be triggered while using some commands together. Would you mind having a look at if this happened in producing or consuming tasks?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@kvrocks.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] memtable, bg_error: Corruption: WriteBatch has wrong count [kvrocks]
Posted by "webfrank (via GitHub)" <gi...@apache.org>.
webfrank commented on issue #2766:
URL: https://github.com/apache/kvrocks/issues/2766#issuecomment-2635866597
Hi, only --rocksdb.block_cache_size 64 (or any other value) just to test memory consumption.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@kvrocks.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] memtable, bg_error: Corruption: WriteBatch has wrong count [kvrocks]
Posted by "git-hulk (via GitHub)" <gi...@apache.org>.
git-hulk commented on issue #2766:
URL: https://github.com/apache/kvrocks/issues/2766#issuecomment-2633501521
@webfrank Is this issue still living after switching to 2.10.1?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@kvrocks.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] memtable, bg_error: Corruption: WriteBatch has wrong count [kvrocks]
Posted by "webfrank (via GitHub)" <gi...@apache.org>.
webfrank commented on issue #2766:
URL: https://github.com/apache/kvrocks/issues/2766#issuecomment-2644997876
@aleksraiden
HSET is around 5kB, some simple keys and one or two binary data (args of the task and result if available) this multiplied by the number of tasks, around 300k
redis-benchmark output:
```
kvrocks@bb0374854df0:/$ redis-benchmark -h 127.0.0.1 -p 6666 -q
WARNING: Could not fetch server CONFIG
PING_INLINE: 21285.65 requests per second, p50=1.103 msec
PING_MBULK: 23820.87 requests per second, p50=1.079 msec
SET: 6615.94 requests per second, p50=5.215 msec
GET: 23534.95 requests per second, p50=1.087 msec
INCR: 6045.22 requests per second, p50=5.879 msec
LPUSH: 5730.99 requests per second, p50=6.143 msec
RPUSH: 5877.86 requests per second, p50=5.215 msec
LPOP: 5724.43 requests per second, p50=5.727 msec
RPOP: 5462.99 requests per second, p50=6.423 msec
SADD: 5952.03 requests per second, p50=5.031 msec
HSET: 6454.11 requests per second, p50=5.471 msec
SPOP: 22946.31 requests per second, p50=1.135 msec
ZADD: 6343.97 requests per second, p50=6.015 msec
ZPOPMIN: 23153.51 requests per second, p50=1.127 msec
LPUSH (needed to benchmark LRANGE): 5858.57 requests per second, p50=5.567 msec
LRANGE_100 (first 100 elements): 18011.53 requests per second, p50=1.439 msec
LRANGE_300 (first 300 elements): 11579.44 requests per second, p50=2.271 msec
LRANGE_500 (first 500 elements): 8394.19 requests per second, p50=3.047 msec
LRANGE_600 (first 600 elements): 7510.33 requests per second, p50=3.479 msec
MSET (10 keys): 5357.05 requests per second, p50=7.431 msec
```
@git-hulk I tried to reproduce it with a script but it only happens using RQ in a random way, usually the first batch of tasks works, the second not.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@kvrocks.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] memtable, bg_error: Corruption: WriteBatch has wrong count [kvrocks]
Posted by "webfrank (via GitHub)" <gi...@apache.org>.
webfrank commented on issue #2766:
URL: https://github.com/apache/kvrocks/issues/2766#issuecomment-2650028903
Hi, I narrowed down the occurrence, it seems that an HSET triggers the corruption, once triggered the system does not work any more and the only way to recover is to start from scratch.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@kvrocks.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] memtable, bg_error: Corruption: WriteBatch has wrong count [kvrocks]
Posted by "webfrank (via GitHub)" <gi...@apache.org>.
webfrank commented on issue #2766:
URL: https://github.com/apache/kvrocks/issues/2766#issuecomment-2662870664
Hi, yes RQ uses MULTI-EXEC a lot but KVRocks handles it. Corruption seems to occur after compaction ends.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@kvrocks.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] memtable, bg_error: Corruption: WriteBatch has wrong count [kvrocks]
Posted by "git-hulk (via GitHub)" <gi...@apache.org>.
git-hulk commented on issue #2766:
URL: https://github.com/apache/kvrocks/issues/2766#issuecomment-2631284272
@webfrank Thanks for your report. Which Kvrocks version you're using?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@kvrocks.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] memtable, bg_error: Corruption: WriteBatch has wrong count [kvrocks]
Posted by "webfrank (via GitHub)" <gi...@apache.org>.
webfrank commented on issue #2766:
URL: https://github.com/apache/kvrocks/issues/2766#issuecomment-2634651954
After "Corruption: WriteBatch has wrong count" the application stops because BLPOP times out. The list is of around 130K elements. I tried manually also BRPOP and it works. It seems something is really corrupted and list pointers are not aligned any more.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@kvrocks.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] memtable, bg_error: Corruption: WriteBatch has wrong count [kvrocks]
Posted by "aleksraiden (via GitHub)" <gi...@apache.org>.
aleksraiden commented on issue #2766:
URL: https://github.com/apache/kvrocks/issues/2766#issuecomment-2635001590
Hmm... In a conspiracy theory - you are use a default options, like:
- rocksdb.enable_pipelined_write no
- rocksdb.delayed_write_rate 0
- rocksdb.write_options.sync no
- rocksdb.write_options.memtable_insert_hint_per_batch no
Or any custom?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@kvrocks.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] memtable, bg_error: Corruption: WriteBatch has wrong count [kvrocks]
Posted by "aleksraiden (via GitHub)" <gi...@apache.org>.
aleksraiden commented on issue #2766:
URL: https://github.com/apache/kvrocks/issues/2766#issuecomment-2639178019
Hmm, I have a test with your options and have no problems. Could you check your instance with redis-benchmark?
E.g.: redis-benchmark -h 127.0.0.1 -p 6666 -q
And how a size of your data, written to HSET?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@kvrocks.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] memtable, bg_error: Corruption: WriteBatch has wrong count [kvrocks]
Posted by "webfrank (via GitHub)" <gi...@apache.org>.
webfrank commented on issue #2766:
URL: https://github.com/apache/kvrocks/issues/2766#issuecomment-2645168741
Usually happens when starting enqueuing tasks or starting new worker. I think some SET command is involved. RQ uses hashsets, list, set and zset.
It would use streams but as kvrocks report version 4 are disabled. Btw, why is version 4 and not 5 if streams are supported?
I will try to find a way to log all those commands using a TCP proxy as following RQ code it is not so simple.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@kvrocks.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org