Hi, I have an issue with the repo.hyron.dev mirror (host id 3010). It is currently marked as inactive due to consecutive crawl failures, but the crawl log is empty, and I'm not sure what the problem is. There was a period of downtime 4 days ago, but it's been up since then, and the content is available under the normal URLs: https://repo.hyron.dev/fedora/linux/releases/40/Workstation/x86_64/iso/ https://repo.hyron.dev/fedora/linux/updates/40/Everything/x86_64/Packages/a/
On 2024-06-03 17:55, Ilya Zlobintsev via Mirror-admin wrote:
Hi, I have an issue with the repo.hyron.dev mirror (host id 3010). It is currently marked as inactive due to consecutive crawl failures, but the crawl log is empty, and I'm not sure what the problem is. There was a period of downtime 4 days ago, but it's been up since then, and the content is available under the normal URLs: https://repo.hyron.dev/fedora/linux/releases/40/Workstation/x86_64/iso/ https://repo.hyron.dev/fedora/linux/updates/40/Everything/x86_64/Packages/a/
Hi. I think the crawler log was just rotated. AFAIK you can just enable your site again.
On the other hand, I had a similar issue back in March: repo.jing.rocks suffered from a kernel lockup caused by NFS for half a day, and fedora's mirror manager just silently disabled the mirror, only then I found out nearly a month later. When I filled out the form in mirror manager, the "email" field says the email will be used for robot notifications, however I've never got any email from bots.
Can fedora please not *silently* disable mirror, not notifying admins? repo.jing.rocks mirrors so many distros that I can't check one by one if it was disabled or not.
Can fedora please not *silently* disable mirror, not notifying admins? repo.jing.rocks mirrors so many distros that I can't check one by one if it was disabled or not.
Agreed, I've opened an enhancement request on MirrorManager about this: https://github.com/fedora-infra/mirrormanager2/issues/342
Hello! I retried the crawl on this mirror and saw this error:
-------- 2024-06-03 13:05:05 INFO repo.hyron.dev (3010): Server load exceeded on https://repo.hyron.dev/fedora/linux/releases/38/Spins/x86_64/iso - trying again in 0.3s (after 1 tries) 2024-06-03 13:05:16 INFO repo.hyron.dev (3010): Server load exceeded on https://repo.hyron.dev/fedora/linux/releases/38/Spins/x86_64/iso - trying again in 0.8s (after 2 tries) 2024-06-03 13:05:31 INFO repo.hyron.dev (3010): Server load exceeded on https://repo.hyron.dev/fedora/linux/releases/38/Spins/x86_64/iso - trying again in 1.4s (after 3 tries) 2024-06-03 13:05:47 INFO repo.hyron.dev (3010): Server load exceeded on https://repo.hyron.dev/fedora/linux/releases/38/Spins/x86_64/iso - trying again in 0.8s (after 4 tries) 2024-06-03 13:05:58 INFO repo.hyron.dev (3010): Server load exceeded on https://repo.hyron.dev/fedora/linux/releases/38/Spins/x86_64/iso - trying again in 9.8s (after 5 tries) 2024-06-03 13:06:18 INFO repo.hyron.dev (3010): Server load exceeded on https://repo.hyron.dev/fedora/linux/releases/38/Spins/x86_64/iso - trying again in 23.2s (after 6 tries) 2024-06-03 13:07:10 INFO repo.hyron.dev (3010): Server load exceeded on https://repo.hyron.dev/fedora/linux/releases/38/Spins/x86_64/iso - trying again in 11.9s (after 7 tries) 2024-06-03 13:07:32 INFO repo.hyron.dev (3010): Server load exceeded on https://repo.hyron.dev/fedora/linux/releases/38/Spins/x86_64/iso - trying again in 107.1s (after 8 tries) 2024-06-03 13:09:30 INFO repo.hyron.dev (3010): Server load exceeded on https://repo.hyron.dev/fedora/linux/releases/38/Spins/x86_64/iso - trying again in 247.6s (after 9 tries) 2024-06-03 13:13:49 INFO repo.hyron.dev (3010): Server load exceeded on https://repo.hyron.dev/fedora/linux/releases/38/Spins/x86_64/iso - giving up after 10 tries --------
The "Server load exceeded" message is logged when there's a HTTP timeout. There's been many successful requests before that, a few that had to be retried but only once or twice. Do you see something relevant in your logs on those timestamps?
Thanks
mirror-admin@lists.fedoraproject.org