[Help] LogRotate returns duplicate records and other

[petrose]

Running on latest TOS7 (7.0.0804) I still getting errors in /var/log/VMs/collectd.log file:
df plugin: Failed to glob vdisk files
Below I have a snippet of the first part where it starts:

Code: Select all

[2026-07-29 11:17:39] Initialization complete, entering read-loop.
[2026-07-29 11:17:39] memory_pool: 创建成功，初始容量 256，最大容量 1024，扩容步长 128
[2026-07-29 11:17:39] memory_pool: 预分配完成，成功分配 64/64 个对象
[2026-07-29 11:17:39] format_json_redis: 内存池已初始化，预分配 64 个 data_entry_t 对象
[2026-07-29 11:17:39] read-function of plugin `virt-0' failed. Will suspend it for 4.000 seconds.
[2026-07-29 11:17:39] df plugin: Failed to glob vdisk files
[2026-07-29 11:17:41] df plugin: Failed to glob vdisk files
[2026-07-29 11:17:43] df plugin: Failed to glob vdisk files
[2026-07-29 11:17:43] virt plugin: Connection established.
[2026-07-29 11:17:45] df plugin: Failed to glob vdisk files
[2026-07-29 11:17:47] df plugin: Failed to glob vdisk files

It persists, and is repeated every 2 seconds !
I reckon there might be some degradation because of this.
Is this fixable, by myself or from the Support side of yours ?

BR Peter

[TMzethar]

I have noted the issue you reported and will verify it. Any further updates will be posted here.

[TMzethar]

Regarding the issue you mentioned, we have fixed it in the newly released version 1.0.057 of VMs.

[rfbjr]

The whole duplicate issue is still there.
It is the way logrorate.conf is coded, it doesn't expect to find xxx.log files already there.

Molly@Forge:/# systemctl status logrotate.service
○ logrotate.service - Rotate log files
Loaded: loaded (/lib/systemd/system/logrotate.service; static)
Active: inactive (dead) since Tue 2026-08-25 08:00:25 EDT; 8min ago
TriggeredBy: ● logrotate.timer
Docs: man:logrotate(8)
man:logrotate.conf(5)
Process: 50879 ExecStart=/usr/sbin/logrotate /tmp/logrotate/logrotate.conf (code=exited, status=1/FAILURE)
Main PID: 50879 (code=exited, status=1/FAILURE)
CPU: 20ms

Aug 25 08:00:25 Forge systemd[1]: Starting Rotate log files...
Aug 25 08:00:25 Forge logrotate[50879]: error: collectd:1 duplicate log entry for /var/log/VMs/collectd.log
Aug 25 08:00:25 Forge logrotate[50879]: error: found error in file collectd, skipping
Aug 25 08:00:25 Forge logrotate[50879]: error: noVNC:1 duplicate log entry for /var/log/noVNC/vnc.log
Aug 25 08:00:25 Forge logrotate[50879]: error: found error in file noVNC, skipping
Aug 25 08:00:25 Forge logrotate[50879]: error: rsync:1 duplicate log entry for /var/log/rsyncd.log
Aug 25 08:00:25 Forge logrotate[50879]: error: found error in file rsync, skipping
Aug 25 08:00:25 Forge systemd[1]: logrotate.service: Deactivated successfully.
Aug 25 08:00:25 Forge systemd[1]: Finished Rotate log files.

[EriChan]

This message is merely a warning and has no impact on system performance or degradation.

Cause:

This happens due to how logrotate.conf handles pre-existing log files, which was not anticipated and causes the duplicate/repeating log entries.

Solution:

You can simply delete /var/log/VMs/collectd.log (or remove the existing log files in that directory). Once deleted, logrotate will handle the files normally and the repeating log messages will stop.

[rfbjr]

This message is merely a warning and has no impact on system performance or degradation.

Cause:

This happens due to how logrotate.conf handles pre-existing log files, which was not anticipated and causes the duplicate/repeating log entries.

Solution:

You can simply delete /var/log/VMs/collectd.log (or remove the existing log files in that directory). Once deleted, logrotate will handle the files normally and the repeating log messages will stop.

Done, Well I did the delete on the first F4-424 Pro before the hourly logrotate run. Ran clean as to be expected. Just deleted the files on the other two devices, so I'll check all three in about 30min.
Thank you