tickets #116084
closedlists.opensuse.org / mailing list web archive - timeouts, sluggishness, nonresponsive, nginx timeout etc
0%
Description
Hello opensuse admin team,
i think i have already reported this kind of situation multiple times:
lists.opensuse.org / mailing list web archive - timeouts, sluggishness,
nonresponsive, nginx timeout etc
sometimes some service were crashed or hanging, but most of my times i
try to simply read the mailing lists via web, they seem slow and
sluggish and often taking minutes until a thread has finished loading if
at all.
which can not be attributed to a completely crashed service or
components i guess.
i already asked if opensuse had enough resources for this webserver and
its components to service the public with mailing list archive for the
web, but i was told that the services and resources were adequate and
not the bottle neck.
thanks.
andreas bittner
Updated by pjessen over 2 years ago
- Category set to Mailing lists
- Private changed from Yes to No
Yes, we have numerous reports on the topic. Currently, I see a uwsgi process running at 100%, it then stopped/completed after about 4min, only to be followed by another one doing the same.
Updated by pjessen over 2 years ago
- Related to tickets #113111: lists.opensuse.org - webserver backand timeouts - nginx 504 gateway timeout added
Updated by pjessen over 2 years ago
Just now received two 502s trying to discard a single held message.
Updated by pjessen over 2 years ago
pjessen wrote:
Just now received two 502s trying to discard a single held message.
Discarding 5-6 messages, one by one, every one gives me a 502.
Updated by pjessen over 2 years ago
- Has duplicate tickets #121993: unable to process held_messages (reject, discard) in mailman added
Updated by crameleon 10 months ago
- Status changed from New to Resolved
- Assignee set to crameleon
Hi,
apologies for the late reply.
Resources are not an issue, but the Mailman software leaking memory, regardless of how much it is granted. In the meanwhile we implemented various improvements.
To summarize what was discussed in and implemented through various smaller tickets:
- There are still issues with the backend software, but automatic memory capping makes incidents not as noticeable lately.
- The nginx performance was improved by offloading redirects to a separate service.
- Archive searching was improved by using a designated search engine service.
Best,
Georg