[slurm-users] Extreme long db upgrade 16.05.6 -> 17.11.3

[Ole Holm Nielsen]

On 07/18/2018 10:56 AM, Roshan Thomas Mathew wrote:
> We ran into this issue trying to move from 16.05.3 -> 17.11.7 with 1.5M 
> records in job table.
> 
> In our first attempt, MySQL reported "ERROR 1206 The total number of 
> locks exceeds the lock table size" after about 7 hours.
> 
> Increased InnoDB Buffer Pool size - 
> https://dba.stackexchange.com/questions/27328/how-large-should-be-mysql-innodb-buffer-pool-size 
> - to 12G (the machine hosting mysql has 128GB) and restarted the 
> conversion and which then completed successfully in 6.5 hours.
> 
> I am sure there are other MySQL tweaks that can be applied catered 
> towards SLURM, will be useful if we can pool them together into the 
> documentation.

I think this is a needle-in-haystack documentation problem :-)

The MySQL optimization has already been documented in 
https://slurm.schedmd.com/accounting.html

I've summarized the information in my Wiki page:
https://wiki.fysik.dtu.dk/niflheim/Slurm_database#mysql-configuration

/Ole