Help Diagnose: Backup error spawn /usr/bin/node EAGAIN

Byte_Smarter

Good morning,

I am hoping some of you guru's here can help me figure this out we had a backup that has been running for over a year without issues just start spitting out this error:

spawn /usr/bin/node EAGAIN

Sep 27 13:42:12 orchestra xo-server[4109]: 2023-09-27T13:42:12.134Z xo:api WARN {redacted} | backupNg.runJob(...) [125ms] =!> Error: spawn /usr/bin/node EAGAIN

I have check syslog and gotten the previous string mentioned and is seems there is no verbosity with Orchestra.

If anyone would be willing to help me start the diagnosis on the issue here I would be most thankful as I don't even know where to begin as it seems like NODE.JS is broken but why would it just break?

Running the following build from source:
Xen Orchestra, commit c714b
xo-server 5.122.0
xo-web 5.124.1

olivierlambert

Hi,

First I would check the prerequisites are OK (Node version first).

edit: also you are not on the latest commits, so as usual, start by being fully up to date first.

Byte_Smarter

@olivierlambert

This is is a development/prod environment so I will have to just pull the latest later today.

I do not feel as that will be the issue as we have a several backups that take place and the one on the 25th ran without issue and saved off 500GB.

node -v v16.20.2

olivierlambert

As per our doc, update to the latest commit before reporting a problem, it might have been fixed already. See the big yellow warning thing here: https://xen-orchestra.com/docs/community.html#report-a-bug
Also as our doc, it says in bold: Please always use latest Node LTS.. Current latest Node LTS is 18, not 16 (18.18.0 LTS as we speak)

Byte_Smarter

@olivierlambert I see, seems like the GPG key for NodeJS apt repo was out, I will get this sorted and then post back this evening / tomo as to if it corrected any issues.

planedrop

What kind of backups are these? Incremental (delta), full, etc..? I think that'll help determining the issue, I had something similar once specifically with an incremental and was able to fix it. May not be related though since that was a while ago and on stable release channel for XOA.