nickva commented on issue #5474:
URL: https://github.com/apache/couchdb/issues/5474#issuecomment-2743780239
Thanks for you report @dianabarsan. I wonder if it's because the downstream
logging service or the pod standard error logger could not cope with the large
amounts of logs being generated and they were being backed up in memory somehow.
With 3.3.3 vs 3.4.2 could be an Erlang VM memory issue that got fixed. It's
not recommended, in general, to run in production with debug logs enabled
persistently.
If you can catch it happening again and have remsh access there are a few
helper functions to pinpoint a particular process that may be consuming memory:
```
couch_debug:resource_hoggers(couch_debug:memory_info(processes()),
heap_size).
| id | heap_size
| code_server[<0.51.0>] | 318187
| erlang:apply/2[<0.701.0>] | 75113
...
```
```
couch_debug:resource_hoggers(couch_debug:memory_info(processes()), binary).
| id | binary
| supervisor_bridge:user_sup/1[<0.71.0>] | 123704
| code_server[<0.51.0>] | 112364
| couch_epi_functions_gen_chttpd_auth[<0.154.0>] | 97656
```
```
couch_debug:resource_hoggers(couch_debug:memory_info(processes()), memory).
| id | memory
| code_server[<0.51.0>] | 6665104
| erlang:apply/2[<0.701.0>] | 2551960
| erlang:apply/2[<0.5439.0>] | 2547712
| config[<0.146.0>] | 1115032
| couch_log_server[<0.180.0>] | 431880
```
```
(node1@127.0.0.1)10>
couch_debug:resource_hoggers(couch_debug:memory_info(processes()),
message_queue_len).
| id | message_queue_len
| init[<0.0.0>] | 0
| erts_code_purger[<0.1.0>] | 0
```
So if you manage to run those we could get some where exactly is the memory
backup. I am guessing it's the log writer process...
--
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: notifications-unsubscr...@couchdb.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
