Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

Redis connection lost and command aborted. code: 'UNCERTAIN_STATE'

See original GitHub issue

We have a node.js app deployed on Heroku and we are using kue to run some background jobs. It usually works fine, but few days back we saw following errors in our production server:

Jul 10 11:19:23 refocus app/web.6: Error removing 4628051 { AbortError: Redis connection lost and command aborted. It might have been processed. 
Jul 10 11:19:23 refocus app/web.6:     at RedisClient.flush_and_error (/app/node_modules/kue/node_modules/redis/index.js:357:23) 
Jul 10 11:19:23 refocus app/web.6:     at RedisClient.connection_gone (/app/node_modules/kue/node_modules/redis/index.js:659:14) 
Jul 10 11:19:23 refocus app/web.6:     at Socket.<anonymous> (/app/node_modules/kue/node_modules/redis/index.js:293:14) 
Jul 10 11:19:23 refocus app/web.6:     at Socket.g (events.js:286:16) 
Jul 10 11:19:23 refocus app/web.6:     at emitNone (events.js:91:20) 
Jul 10 11:19:23 refocus app/web.6:     at Socket.emit (events.js:185:7) 
Jul 10 11:19:23 refocus app/web.6:     at endReadableNT (_stream_readable.js:975:12) 
Jul 10 11:19:23 refocus app/web.6:     at _combinedTickCallback (internal/process/next_tick.js:74:11) 
Jul 10 11:19:23 refocus app/web.6:     at process._tickDomainCallback [as _tickCallback] (internal/process/next_tick.js:122:9) 
Jul 10 11:19:23 refocus app/web.6:   code: 'UNCERTAIN_STATE', 
Jul 10 11:19:23 refocus app/web.6:   command: 'EXEC', 
Jul 10 11:19:23 refocus app/web.6:   errors:  
Jul 10 11:19:23 refocus app/web.6:    [ { AbortError: Redis connection lost and command aborted. It might have been processed. 
Jul 10 11:19:23 refocus app/web.6:          at RedisClient.flush_and_error (/app/node_modules/kue/node_modules/redis/index.js:357:23) 
Jul 10 11:19:23 refocus app/web.6:          at RedisClient.connection_gone (/app/node_modules/kue/node_modules/redis/index.js:659:14) 
Jul 10 11:19:23 refocus app/web.6:          at Socket.<anonymous> (/app/node_modules/kue/node_modules/redis/index.js:293:14) 
Jul 10 11:19:23 refocus app/web.6:          at Socket.g (events.js:286:16) 
Jul 10 11:19:23 refocus app/web.6:          at emitNone (events.js:91:20) 
Jul 10 11:19:23 refocus app/web.6:          at Socket.emit (events.js:185:7) 
Jul 10 11:19:23 refocus app/web.6:          at endReadableNT (_stream_readable.js:975:12) 
Jul 10 11:19:23 refocus app/web.6:          at _combinedTickCallback (internal/process/next_tick.js:74:11) 
Jul 10 11:19:23 refocus app/web.6:          at process._tickDomainCallback [as _tickCallback] (internal/process/next_tick.js:122:9) 
Jul 10 11:19:23 refocus app/web.6:        code: 'UNCERTAIN_STATE', 
Jul 10 11:19:23 refocus app/web.6:        command: 'ZREM', 
Jul 10 11:19:23 refocus app/web.6:        args: [Object], 
Jul 10 11:19:23 refocus app/web.6:        position: 0 },  .............

We have seen these ‘UNCERTAIN_STATE’ errors before as well. I went through other related issues and it seems a network problem, but since it looks like it is coming from kue, I just wanted to confirm whether we have some potential options to prevent this in future.

Issue Analytics

State:
Created 6 years ago
Reactions:1
Comments:11

Top GitHub Comments

3reactions

mzalazarcommented, Jun 12, 2018

I have found a limit with my server (max connextion): root@ar:/var/log# sysctl net.core.somaxconn net.core.somaxconn = 128 Using this command, i changed my settings sysctl -w net.core.somaxconn=1024 (you can “tune” this setting) By using: sysctl -p we apply new settings to kernel… and i had to restart redis of course.

2reactions

danielmhanovercommented, Mar 18, 2019

Also getting this issue. Is it a connections count issue?