Getting back the original question - I just noticed that there is special
option AuditRPCs in DebugFlags for controller, so perhaps you can determine
source of RPC calls without breaking things.
https://slurm.schedmd.com/slurm.conf.html#OPT_AuditRPCs
Regards
Patryk.
On 25/05/07 10:47AM, Patryk
> IMHO the RPC rate limiting should be considered a best practice, and I
> wouldn't think that it's a "dirty" configuration. You need Slurm 23.02 or
> later for this. Some details are discussed in this Wiki page:
Dirty in a way that levels are so low that they break some other service in
order t
ys as documented in
https://wiki.fysik.dtu.dk/Niflheim_system/Slurm_database/#setting-maxquerytimerange
Best regards,
Ole
- Mail original -
De: "Ole Holm Nielsen via slurm-users"
À: slurm-users@lists.schedmd.com
Envoyé: Mercredi 7 Mai 2025 10:13:08
Objet: [slurm-users] Re: Lots o
: "Ole Holm Nielsen via slurm-users"
À: slurm-users@lists.schedmd.com
Envoyé: Mercredi 7 Mai 2025 10:13:08
Objet: [slurm-users] Re: Lots of RPC calls and REQUEST_GETPW calls
On 5/7/25 09:57, Patryk Bełzak via slurm-users wrote:
> Hi,
> why you think it's an authentication
On 5/7/25 09:57, Patryk Bełzak via slurm-users wrote:
Hi,
why you think it's an authentication requests? As far as I understand multiple
UIDs are asking for job and partition info. It's unlikely that all of them
perform that kind of requests the same way and in the same time, so I think you
sh
Hi,
why you think it's an authentication requests? As far as I understand multiple
UIDs are asking for job and partition info. It's unlikely that all of them
perform that kind of requests the same way and in the same time, so I think you
should look for some external program that may do that - i
