Opened 5 years ago
Closed 5 years ago
#2433 closed task (fixed)
old-web: irqbalance: WARNING, didn't collect load info for all cpus, balancing is broken
| Reported by: | strk | Owned by: | |
|---|---|---|---|
| Priority: | normal | Milestone: | Sysadmin Contract 2020-I |
| Component: | SysAdmin | Keywords: | |
| Cc: |
Description
I found many such lines in /var/log/syslog, sounds like something to fix:
Apr 7 06:52:42 old-web /usr/sbin/irqbalance: WARNING, didn't collect load info for all cpus, balancing is broken Apr 7 06:52:52 old-web /usr/sbin/irqbalance: WARNING, didn't collect load info for all cpus, balancing is broken Apr 7 06:53:02 old-web /usr/sbin/irqbalance: WARNING, didn't collect load info for all cpus, balancing is broken Apr 7 06:53:12 old-web /usr/sbin/irqbalance: WARNING, didn't collect load info for all cpus, balancing is broken Apr 7 06:53:22 old-web /usr/sbin/irqbalance: WARNING, didn't collect load info for all cpus, balancing is broken Apr 7 06:53:32 old-web /usr/sbin/irqbalance: WARNING, didn't collect load info for all cpus, balancing is broken Apr 7 06:53:42 old-web /usr/sbin/irqbalance: WARNING, didn't collect load info for all cpus, balancing is broken Apr 7 06:53:52 old-web /usr/sbin/irqbalance: WARNING, didn't collect load info for all cpus, balancing is broken Apr 7 06:54:02 old-web /usr/sbin/irqbalance: WARNING, didn't collect load info for all cpus, balancing is broken Apr 7 06:54:12 old-web /usr/sbin/irqbalance: WARNING, didn't collect load info for all cpus, balancing is broken
Change History (3)
comment:1 by , 5 years ago
comment:2 by , 5 years ago
I tried rebooting but that didn't make the message go away. Stuff I've read suggests this service may not be needed particularly in a virtual environment. So I stopped the service for now.
comment:3 by , 5 years ago
| Resolution: | → fixed |
|---|---|
| Status: | new → closed |
going to close this and reopen if shutting the service caused any issues.
Note:
See TracTickets
for help on using tickets.
I think I need to reboot osgeo7 since the lxd 4 upgrade -- doing things like lxc list are slow.
I suspect it will fix this issue as well, though maybe rebooting old-web might be sufficient as I think that might be one where I had to unmount the proc things and I might not have restarted after. I'll do that first.