Sometimes, some kvms are having very very high IO like 1-2 G Disk IO:
And that makes vm unresponsive. So is this related to Proxmox / VM itself or Zfs file system?
Thats seems to be impossible inside vm but these vms have sql server installed. So where should go first you think?
I am running influxdb in a nat of pve kvm environment.
I see randomly:
metrics send error 'Monitor': 500 Can't connect to 8.8.8.8:8086 (Connection timed out) but metrics are being collected.
I have forwarded 8086 to influxdb internal nat ip (in pve).
What could be issue?
Today, there was an issue on nas, pbs and metric servers and unreachable for a while (all are on same hardware) then all proxmox servers had issues those were trying to connect to it. Like disks status are unavaliable. Gui is not running correctly. Vm names were missing etc.
We tried all...
I have several nat moded vms in pve. I also installed pbs like this. So fas so good. But I noted today, it can not reach internet.
Is there anything I need to allow? On same Pve, other vms are fine with internet, but pbs can not reach internet with internal ip (192.168.1.10).
It can reach...
Do I need to create full backup regularly? I do not see such option. But AFAIK, when using incremental backups, time to time full backups should be taken. Is this same with pbs? Or I can just leave it one full backup and let rests for incrementals?
How can I keep second copy of backups to an other disk. Sync job ask for pbs. I install pbs in a pve vm. And attached disk to it. I have some more disk free. So I think of keeping copy of disk in use with pbs. Or any better advise?
Or should I install an other pbs on same pve not and use it...
Our Pve servers has ssd sata for only pve os. But we think of using normal sata disk (7200rpm) . Do you think if would impact on overall pve performance? Vms are running on nvme disks. We think Sata disks will last longer because ssds may be failing in any time.
AFIK, metric does not provide cpu temps. After some search, I read k10temp module is a workaround. https://github.com/groeck/k10temp
So is it possible to install that module ? Or anyway to get cpu temp with "sensors -j" ?
When externalsender.com sends an email to a user behind PMG, he gets following spf error.
And of course he is not allowed to send mail from PMG's ip. Why is this happening?
<user@domain-behind-pmg.com>: host domain-behind-pmg.com[1.2.3.4 > (User's server ip)]
said:
550-SPF: 2.3.4.5> (PMG's...
When I check memory usage in windows, total memory in use does not sound correct. Please see screen shot, that vm has total 20 GB Ram but windows show %85 in use which means 17 GB but if you look at processes it is not in total 17 GB. Maybe around 8-9 GB.
Using zfs..
What is the difference of using builtin rulesets from gui or spamassian rules custom /etc/mail/spamassassin/custom.cf ? Performance / usage / easy to manage?
I loose connection in server and have guys in DC to rebooted server. Then when I check logs only thing I see at connection lost time:
Oct 31 12:00:20 s8 CRON[911689]: pam_unix(cron:session): session closed for user root
-- Reboot --
Oct 31 12:22:11 s8 kernel: Linux version 5.15.64-1-pve...
One node logs this error and vms are not reachable:
VM 221 qmp command failed - VM 221 qmp command 'query-proxmox-support' failed - unable to connect to VM 221 qmp socket - timeout after 31 retries
Could this be vm is eating so much resource or something related to pve? We have all updates...
Ok when adding transport, there is a "Use Mx" option.
Lets say domain.com has two mx records (second is for fallback):
1- pmg.domain.com Priority 0
2. mail.domain.com Priority 10
Then I add transport hostname as domain.com (not sure if I must enter mail.domain.com ? ) and check Use Mx option...
On windows logs I found:
The computer has rebooted from a bugcheck. The bugcheck was: 0x0000003b (0x00000000c0000005, 0xfffff8000a8c4ce3, 0xffff848f31524a40, 0x0000000000000000). A dump was saved in: C:\WINDOWS\MEMORY.DMP. Report Id: 34f4e739-31a2-446c-b6ab-997ef65308c5.
proxmox-ve: 7.2-1...
I tried all methods in forum to apply network changes like adding a new nat ip to interfaces file OR new port forwaring without rebooting the node. But I got no stabil results. And at the end I had to reboot node.
So this made me think if I can create a wm for only these nat thing just for no...
I am currently using a pve as a backup server for all my zfs server with pve-zsync. And when I need, I can restore / mount dataset and get file from it.
Can I do this with pbs too ? Can I also create vm with pbs?
I want to use a wget command to download a file and see network speed test for pve server from a remote pc.
So I added a test.zip file to /usr/share/pve-manager/images and used browsers to see http://s1.hostname.com:8006/pve2/images/test.zip
But getting 501 error.
Is there a better or easier...
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.