proxmox crashes
- Thread starter jhmc93
- Start date
Cephfs? I was uploading to my local disk on the local storage on the proxmox gui, I’m using all the latest stuff as it’s a new install with proxmox 8.0.4
Thanks, I had hard crashes recently uploading to cephfs volume while running a later linux kernel, your issue is not that.
Without seeing the top of that crash report not sure it's possible to even guess what the issue is. Does the same happen on all your cluster nodes?
Without seeing the top of that crash report not sure it's possible to even guess what the issue is. Does the same happen on all your cluster nodes?
Last edited:
this is a test pc so its only running on one machine at the moment, i did apt update and apt upgrade and made sure everything was up to date. it was and i tried uploading to the local volume again. Can you tell me how i get a log of the crashes as it has crashed again.
i haven't figured that out (or rather i know it may need kdump and or setting up serial connection to redirect stderr to - but how is another matter) - when it panics can you take a picture of the top of the screen - that's the first place to start, you cut off the top of the screen... and you uploaded with wrong rotation too...
Last edited:
unfortunately i think it's a net problem as per syslog:
if i'm right i'am not sure whats kicking it off
i don't think that has much do to do with your crash
those messages are perfectly normal
so are the postfix messages unless they are happening with high frequency - for example i have lot of the same messages back on the 23rd
the last entries on the initial screenshot say 'tg3' i believe thats a broadcom nic module - if so maybe you network driver was the issue?
https://wiki.archlinux.org/title/Network_configuration/Ethernet
tl;dr if you have seen no more crashes and you see no explicit errors in dmesg - move on and put it down to one of those things
those messages are perfectly normal
so are the postfix messages unless they are happening with high frequency - for example i have lot of the same messages back on the 23rd
the last entries on the initial screenshot say 'tg3' i believe thats a broadcom nic module - if so maybe you network driver was the issue?
https://wiki.archlinux.org/title/Network_configuration/Ethernet
tl;dr if you have seen no more crashes and you see no explicit errors in dmesg - move on and put it down to one of those things
All I know is everytime I turn it on start a vm or upload a iso it disconnects from my internet, I know it using a IP that’s not in use somewhere else, I think it must be a network driver issue because as soon as I run systemctl restart networking it returns to normal
Yes, that would be true, but it would allow people to understand better what they are looking at. While the defaults are present, as you have indicated, we don't know all of the interfaces and how they have been configured; it helps us help you.
here's the network page, i also included DNS page and network interfaces in the terminal
Attachments
Okay, jump to the section "Setting the Speed and Duplex" in the following doc:
https://wiki.debian.org/NetworkConfiguration
First, test the negotiated speeds as indicated, then try setting a lower speed manually, like 100Mbps. See if it improves the situation; this is only as a test.
Do you have a different ethernet cable and switch you can test? [This is one of those problems where you'll have to work through everything, just in case. Faulty cables can be a true nightmare.]
If you can add a network card, for testing purposes, you'll be able to see if the machine suddenly behaves when a VM is spun up.
With all the above, it's a case of establishing if there is some kind of negotiation issue with the driver/NIC, and also if it is the NIC at all; using a different network card should nail that down.
https://wiki.debian.org/NetworkConfiguration
First, test the negotiated speeds as indicated, then try setting a lower speed manually, like 100Mbps. See if it improves the situation; this is only as a test.
Do you have a different ethernet cable and switch you can test? [This is one of those problems where you'll have to work through everything, just in case. Faulty cables can be a true nightmare.]
If you can add a network card, for testing purposes, you'll be able to see if the machine suddenly behaves when a VM is spun up.
With all the above, it's a case of establishing if there is some kind of negotiation issue with the driver/NIC, and also if it is the NIC at all; using a different network card should nail that down.
the only way I can add another Network interface is through usb 3.0 port as it's one of these machines:
