Note creation hanging with message 'saving notes to database..'

[tlukasha]

I have one database with about 250,000 hands where note creation works fine.

I have added another database on the same machine where I have about 100k hands. On this database I was able to create all the notes on the first occasion. I then changed some definitions, reset all notes, and wanted to re-create, but now it gets stuck about 40k hands in.

I have re-tested on the other database and on there it still completes the process fine after a reset.

Any idea how I can debug why it is seemingly hanging? Hand importing is off atm, but I can see that holdemmanager.exe is taking 5-10% cpu in task manager, so it is clearly doing something. Just seems weird that it should suddenly take 30 minutes plus to do a save... The machine is powerful, plenty of available cpu, lots of memory, etc. I am running the 64 bit version of postgres, but I have done so without problems for 3+ years already.

[tlukasha]

Update on this... after an hour or so of showing 'saving to database' it continued on and made its way to the end of the processing of the 100k hands, however, again it gets stuck on the save operation at the end.

After waiting for a long time I restarted HM2 and it restarted the processing around the 50k hand mark, made its way to the end but again never completed the saving operation.

[tlukasha]

Update on this... after an hour or so of showing 'saving to database' it continued on and made its way to the end of the processing of the 100k hands, however, again it gets stuck on the save operation at the end.

After waiting for a long time I restarted HM2 and it restarted the processing around the 50k hand mark, made its way to the end but again never completed the saving operation.

For what it is worth, I can see in the resource monitor that Postgres is writing to disk, but for some reason an operation that should take 10 minutes is taking multiple hours and in fact does not seem to complete at all.

[tlukasha]

For what it is worth, I can see in the resource monitor that Postgres is writing to disk, but for some reason an operation that should take 10 minutes is taking multiple hours and in fact does not seem to complete at all.

The process did not complete overnight, but took 4-8 hours rather than the 4-8 minutes it took the previous day. What could cause such extreme degradation in performance?

[tlukasha]

The process did not complete overnight, but took 4-8 hours rather than the 4-8 minutes it took the previous day. What could cause such extreme degradation in performance?

It 'DID' complete overnight I attempted to say..

[sreticentv]

Which version of windows are you running? Windows 8+ lets you see how much of your drive drive bandwidth is being used via task manager. My guess is that should be hovering close to 100% during stalls like that. If not then there is likely some issue with hand histories causing a lot of errors. You can have a look at your notecaddy error logs which you'd find in

C:\Users\{windowslogin}\AppData\Roaming\HoldemMana ger\NoteCaddyLogs

If there is/are huge error log files then that's almost certainly the cause