worker=2
Detected dead worker 73, restarting...
I, [2015-10-21T22:12:58.139406 #7622] INFO -- : worker=1 ready
I, [2015-10-21T22:12:58.139752 #7626] INFO -- : worker=2 ready
Loading Sidekiq in process id 7632
D, [2015-10-21T22:13:27.161680 #50] DEBUG -- : waiting 16.0s after suspend/hibernation
I, [2015-10-22T07:35:12.745462 #50] INFO -- : master reopening logs...
I, [2015-10-22T07:35:12.812351 #142] INFO -- : worker=0 reopening logs...
I, [2015-10-22T07:35:12.817665 #7622] INFO -- : worker=1 reopening logs...
I, [2015-10-22T07:35:12.821608 #7626] INFO -- : worker=2 reopening logs...
```
---
### Author:  [@mpalmer](https://meta.discourse.org/u/mpalmer)
#### Post date: [26 באוקטובר, 2015, 9:51pm UTC](https://meta.discourse.org/t/unicorn-log-files-using-up-all-the-space-on-my-box/34937/6 "2015-10-26T21:51:44Z")
Yes, we have log rotation configured, and it did work in this case – the huge file wasn’t the one “currently” in use, but rather was rotated a few days ago. _Something_ was causing Unicorn to puke up huge amounts of error logs for less than a day, and then it stopped (could have been because the disk filled up, I guess). Without knowing what it was, it’s hard to give a solid diagnosis, but if the problem isn’t happening now, I wouldn’t worry about it. If it starts happening again, keep the logs!
---
### Author:  [@bioball](https://meta.discourse.org/u/bioball)
#### Post date: [26 באוקטובר, 2015, 10:14pm UTC](https://meta.discourse.org/t/unicorn-log-files-using-up-all-the-space-on-my-box/34937/7 "2015-10-26T22:14:29Z")
If that’s the case, then it’s kind of alarming that logging so much in one day can use up all my disk space. It brought forum down, and nginx started responding with 502’s. Perhaps you guys can also rotate the logs based on file size, to avoid these kinds of situations?
---
### Author:  [@codinghorror](https://meta.discourse.org/u/codinghorror)
#### Post date: [26 באוקטובר, 2015, 10:22pm UTC](https://meta.discourse.org/t/unicorn-log-files-using-up-all-the-space-on-my-box/34937/8 "2015-10-26T22:22:37Z")
Like I said, there was something catastrophically wrong with your specific install. Never seen this happen before across hundreds of installs..
If we had the log file, we could look, but.. we don’t
---
### Author:  [@bioball](https://meta.discourse.org/u/bioball)
#### Post date: [26 באוקטובר, 2015, 10:23pm UTC](https://meta.discourse.org/t/unicorn-log-files-using-up-all-the-space-on-my-box/34937/9 "2015-10-26T22:23:49Z")
Ok… interesting.
I’ll report back if this happens again. I’m planning on re-doing the install.
---
### Author:  [@mpalmer](https://meta.discourse.org/u/mpalmer)
#### Post date: [26 באוקטובר, 2015, 11:20pm UTC](https://meta.discourse.org/t/unicorn-log-files-using-up-all-the-space-on-my-box/34937/10 "2015-10-26T23:20:57Z")
> [@bioball](#):
>
> Perhaps you guys can also rotate the logs based on file size, to avoid these kinds of situations?
Logrotate still only runs once a day. Size-based rotation wouldn’t have saved you here. Perhaps you could monitor your server for free space, so you could catch these kinds of situations?
---
### Author:  [@bioball](https://meta.discourse.org/u/bioball)
#### Post date: [27 באוקטובר, 2015, 11:25pm UTC](https://meta.discourse.org/t/unicorn-log-files-using-up-all-the-space-on-my-box/34937/11 "2015-10-27T23:25:11Z")
I’m not too familiar with logrotate, as I don’t use it, but it seems like you might be able to configure it to run [hourly](http://stackoverflow.com/questions/25485047/hourly-rotation-of-files-using-logrotate). I rotate logs on my own servers based on size, but I’ve always used my own logging module within my server rather than a separate process.
Anyway. If this has never happened to anybody else before I can understand that it’s not a big deal, but I will report back if it does happen to me again, plus whatever was being logged out.
---
### Author:  [@JonLumb](https://meta.discourse.org/u/JonLumb)
#### Post date: [4 באפריל, 2016, 10:14am UTC](https://meta.discourse.org/t/unicorn-log-files-using-up-all-the-space-on-my-box/34937/12 "2016-04-04T10:14:34Z")
Ok, I’ve got the same issue occurring on my Discourse install. I followed the Docker install instructions originally and everything seemed to be fine, but have started having this trouble recently. I did have several beta updates go through that fell over part way through and I had to cancel.
Is doing an upgrade to the latest version likely to resolve this problem (I’ll be doing it via SSH as that’s always been successful for me), or am I looking at backing up my config file and reinstalling?
---
### Author:  [@codinghorror](https://meta.discourse.org/u/codinghorror)
#### Post date: [4 באפריל, 2016, 6:30pm UTC](https://meta.discourse.org/t/unicorn-log-files-using-up-all-the-space-on-my-box/34937/13 "2016-04-04T18:30:38Z")
What does the unicorn log file actually say, if it is so large?
---
### Author:  [@JonLumb](https://meta.discourse.org/u/JonLumb)
#### Post date: [7 באפריל, 2016, 5:22am UTC](https://meta.discourse.org/t/unicorn-log-files-using-up-all-the-space-on-my-box/34937/14 "2016-04-07T05:22:08Z")
In short, I’m struggling to find out as nano / vi won’t actually open it due to it being so big.
---
### Author:  [@tgxworld](https://meta.discourse.org/u/tgxworld)
#### Post date: [7 באפריל, 2016, 5:26am UTC](https://meta.discourse.org/t/unicorn-log-files-using-up-all-the-space-on-my-box/34937/15 "2016-04-07T05:26:16Z")
Try [tailing](https://en.wikipedia.org/wiki/Tail_(Unix)) the file?
---
### Author:  [@JonLumb](https://meta.discourse.org/u/JonLumb)
#### Post date: [7 באפריל, 2016, 5:29am UTC](https://meta.discourse.org/t/unicorn-log-files-using-up-all-the-space-on-my-box/34937/16 "2016-04-07T05:29:24Z")
I’ve just deleted the log again in order to get the forum back up, but it seems to suddenly leap in size about once every 2 days, so I’ll do that next time. Thanks for the tip!
---
### Author:  [@twifno](https://meta.discourse.org/u/twifno)
#### Post date: [19 באפריל, 2016, 4:00pm UTC](https://meta.discourse.org/t/unicorn-log-files-using-up-all-the-space-on-my-box/34937/17 "2016-04-19T16:00:14Z")
I have also encountered this issue, and it seems caused by out-of-memory. 1G memory + 2G SWAP seems not enough for hosting discourse, I have upgraded to 2G though.
the tail of the error log:
```plaintext
E, [2016-04-19T07:53:47.411276 #54] ERROR -- : /var/www/discourse/vendor/bundle/ruby/2.0.0/gems/activesupport-4.2.6/lib/active_support/core_ext/kernel/agnostics.rb:7:in ``'
E, [2016-04-19T07:53:47.411301 #54] ERROR -- : config/unicorn.conf.rb:84:in `max_rss'
E, [2016-04-19T07:53:47.411325 #54] ERROR -- : config/unicorn.conf.rb:99:in `out_of_memory?'
E, [2016-04-19T07:53:47.411349 #54] ERROR -- : config/unicorn.conf.rb:122:in `check_sidekiq_heartbeat'
E, [2016-04-19T07:53:47.411381 #54] ERROR -- : config/unicorn.conf.rb:146:in `master_sleep'
E, [2016-04-19T07:53:47.495358 #54] ERROR -- : /var/www/discourse/vendor/bundle/ruby/2.0.0/gems/unicorn-5.0.1/lib/unicorn/http_server.rb:284:in `join'
E, [2016-04-19T07:53:47.495412 #54] ERROR -- : /var/www/discourse/vendor/bundle/ruby/2.0.0/gems/unicorn-5.0.1/bin/unicorn:126:in `'
E, [2016-04-19T07:53:47.495464 #54] ERROR -- : /var/www/discourse/vendor/bundle/ruby/2.0.0/bin/unicorn:22:in `load'
E, [2016-04-19T07:53:47.495489 #54] ERROR -- : /var/www/discourse/vendor/bundle/ruby/2.0.0/bin/unicorn:22:in `'
I, [2016-04-19T07:53:47.495527 #54] INFO -- : master reopening logs...
```
---
### Author:  [@JonLumb](https://meta.discourse.org/u/JonLumb)
#### Post date: [24 באפריל, 2016, 8:47pm UTC](https://meta.discourse.org/t/unicorn-log-files-using-up-all-the-space-on-my-box/34937/18 "2016-04-24T20:47:03Z")
Right, finally had it explode when I had the time to do some more diagnostics, piped the last 5000 lines from unicorn.stderr.log to a new file:
[unicornsample.txt](https://global.discourse-cdn.com/meta/original/3X/5/9/5919dd507c6150df7ec4f27d7d9ef0456864f5f6.txt) (443.8 KB)
Failing there being anything obvious in the error log, is there a straightforward way to reinstall the software without touching the database and the config file?
---
### Author:  [@JonLumb](https://meta.discourse.org/u/JonLumb)
#### Post date: [25 באפריל, 2016, 11:44am UTC](https://meta.discourse.org/t/unicorn-log-files-using-up-all-the-space-on-my-box/34937/19 "2016-04-25T11:44:30Z")
Hmm, from what I can see in the logs it looks like it might be related to Redis, but the instructions I can see for reinstalling that are definitely not part of what I did to set up Discourse!
---
### Author:  [@system](https://meta.discourse.org/u/system)
#### Post date: [11 בינואר, 2019, 12:03pm UTC](https://meta.discourse.org/t/unicorn-log-files-using-up-all-the-space-on-my-box/34937/20 "2019-01-11T12:03:26Z")
[Next page](https://meta.discourse.org/t/unicorn-log-files-using-up-all-the-space-on-my-box/34937.md?page=2)