Project

General

Profile

Actions

tickets #92966

closed

MirrorBrain scanner is stuck

Added by quantummirror almost 3 years ago. Updated almost 2 years ago.

Status:
Resolved
Priority:
Normal
Assignee:
Category:
Mirrors
Target version:
-
Start date:
2021-05-22
Due date:
% Done:

100%

Estimated time:

Description

Dear openSUSE infra team,

I think the MB scanner is stuck and it does not update the file lists. I
also see the normal scanner connections from 195.135.221.170 on opensuse
dir and on the isos.

My mirror hasn't received a single iso or package (ZYpp) download since
yesterday morning. [21/May/2021:08:56:32 +0200]

Output from
(h)ttps://download.opensuse.org/tumbleweed/iso/openSUSE-Tumbleweed-DVD-x86_64-Snapshot20210519-Media.iso?mirrorlist

"No mirror was found

I am very sorry, but no mirror was found. Feel free to download directly:
/tumbleweed/iso/openSUSE-Tumbleweed-DVD-x86_64-Snapshot20210519-Media.iso
https://download.opensuse.org/tumbleweed/iso/openSUSE-Tumbleweed-DVD-x86_64-Snapshot20210519-Media.iso

Powered by MirrorBrain" http://mirrorbrain.org/

The other download pages also give this result.

Would you be so kind and fix this issue?

Thank you for your help in advance!

Cheers,

Peter

Actions #1

Updated by cboltz almost 3 years ago

  • Category set to Mirrors
  • Assignee set to pjessen
Actions #2

Updated by pjessen almost 3 years ago

  • Private changed from Yes to No

Bernhard already looked at it and determined it was mirrordb2 having a problem:

I fixed it for now my switching pontifex2:
/etc/apache2/vhosts.d/_download.conf
to use mirrordb1 again.

mirrordb2's postgres would not start (probably replication needs some
kicking again) but I wonder if that replication is really worth all the hassle.

Often enough HA setups decrease availability via increased complexity
causing increased fragility.

E.g. booting up a single node always just works, but rebooting a pair of
HA nodes might well have both of them reject work for fear of split-brain.

The scanner continues to use mirrordb1.

Actions #3

Updated by bmwiedemann almost 3 years ago

I also setup a local haproxy on pontifex2 to forward to mirrordb2 if available and fallback to mirrordb1 otherwise.

The advantage is that this makes outages less noticeable, but the downside is that we also notice outages less, so would need better monitoring.

Actions #4

Updated by pjessen over 2 years ago

  • Assignee deleted (pjessen)
Actions #5

Updated by pjessen almost 2 years ago

  • Assignee set to andriinikitin
Actions #6

Updated by andriinikitin almost 2 years ago

  • Status changed from New to Resolved
  • % Done changed from 0 to 100

This should be resolved already, so I closing the call.
Let us know if any issue.

Actions

Also available in: Atom PDF