The error shows up at https://MyServerAddress.com:8007/#pbsServerAdministration:tasks, during KVM restore.
On the other hand, in proxmox server https://MyServerAddress.com:8006/#v1:0:18:4:::::::, the restore task is completed without error.
I’am running pve-manager/6.2-11/22fb4983 (running kernel: 5.4.65-1-pve) with proxmox-backup-server 0.8.19-1
This is a similar behaviour with bug reported here, meaning there is no negative impact upon restoring process :
Possible related issue: https://forum.proxmox.com/threads/error-when-finishing-task-of-backup.76379/#post-339815
Thx!
Code:
…
2020-09-23T00:39:06+03:00: GET /chunk
2020-09-23T00:39:06+03:00: download chunk "/ab/store_melania/.chunks/7b76/7b76c3f2a082015fbf28719166b4e470def1063339f653b9dfb5892b71917f03"
2020-09-23T00:39:06+03:00: GET /chunk
2020-09-23T00:39:06+03:00: download chunk "/ab/store_melania/.chunks/63c1/63c1eec60c1da3550078c4240ab986dcfb9a0809ae2e5c51715442ead02a48d6"
2020-09-23T00:39:06+03:00: GET /chunk
2020-09-23T00:39:06+03:00: download chunk "/ab/store_melania/.chunks/b5e8/b5e88aa07f327df0b9440e59c42fd7f08207c78cf9a0007ae0033f45b46b0e81"
2020-09-23T00:39:06+03:00: TASK ERROR: connection error: Transport endpoint is not connected (os error 107)
On the other hand, in proxmox server https://MyServerAddress.com:8006/#v1:0:18:4:::::::, the restore task is completed without error.
Code:
progress 100% (read 2361393152 bytes, zeroes = 14% (331350016 bytes), duration 46 sec)
restore image complete (bytes=2361393152, duration=46.62s, speed=48.30MB/s)
rescan volumes...
TASK OK
I’am running pve-manager/6.2-11/22fb4983 (running kernel: 5.4.65-1-pve) with proxmox-backup-server 0.8.19-1
This is a similar behaviour with bug reported here, meaning there is no negative impact upon restoring process :
It turns out that this log message will be emitted every time a forwarded connection completes. From the proxy's point of view, the socket failed. Why it failed, it can't really know (aside from the errno, which in this case indicates "Transport endpoint is not connected").
So, if you're not actually observing any functional problems in your application, my suggestion would be to simply reduce the log level of these messages to DEBUG so that they don't show up in logs by default. Would that be sufficient?
Possible related issue: https://forum.proxmox.com/threads/error-when-finishing-task-of-backup.76379/#post-339815
Thx!