So the only mention of this I can find is a "rollup" mention in https://slurm.schedmd.com/slurmdbd.conf.html , now that I specificaly googled for "slurmdbd rollup" .
So if this hourly summary is the right behaviour, I'd request it be better documented -- nothing at all is mentioned in https://slurm.schedmd.com/accounting.html Thanks. On Sun, Oct 14, 2018 at 02:18:48PM -0700, Douglas Jacobsen wrote: > Sreport shows data that is summarized hourly. Restarting slurmdbd can delay > this process. If some jobs are missing end records it can massively slow > the process because it may need to pick a much earlier start time in the > past to summarize. > > Sacctmgr show runawayjobs can help identify if you are in this situation > > On Sun, Oct 14, 2018 at 2:05 PM Dave Botsch <bot...@cnf.cornell.edu> wrote: > > > This seems to reflect what I am seeing. Someone earlier mentioned > > multiple restarts of slurmdbd... those restarts never made data appear > > unless right around on the hour. > > > > It's as if instead of data getting sent right through slurmdbd that > > something in slurmdbd is just doing an hourly check of the text based > > sacct records (which I don't understand why those are even there if not > > configured in slurm.conf). > > > > Thanks. > > > > On Sun, Oct 14, 2018 at 12:08:10PM +0100, Antony Cleave wrote: > > > I have noticed on several clusters that sreport can be upto one hour out > > of > > > date i.e. it will update on the hour every hour. > > > > > > sacct does not behave this way and is always up to date. > > > > > > I cannot see this stated in the docs or see any config settings to > > control > > > this but it happens on the last 17.02 cluster I checked. > > > > > > Antony > > > > > > On 14 Oct 2018 11:58, "Steven Dick" <kg4...@gmail.com> wrote: > > > > > > It is documented that you need to create the cluster in the database. > > > > > > It is not documented that the accounting system won't work until you > > > restart slurmdbd multiple times before it starts collecting accounting > > > records. > > > > > > Also, none of the necessary restarts are needed on an upgrade -- only > > > when slurm is initialized for a new cluster. > > > > > > > > > On Sun, Oct 14, 2018 at 4:12 AM Ole Holm Nielsen > > > <ole.h.niel...@fysik.dtu.dk> wrote: > > > > Correct, and this is documented in the Slurm accounting setup page: > > > > https://slurm.schedmd.com/accounting.html#database-configuration > > > > > > > > -- > > ******************************** > > David William Botsch > > Programmer/Analyst > > @CNFComputing > > bot...@cnf.cornell.edu > > ******************************** > > > > -- > Sent from Gmail Mobile -- ******************************** David William Botsch Programmer/Analyst @CNFComputing bot...@cnf.cornell.edu ********************************
