[GENERAL] Tablespaces across same raid...

i just have one raid controller with 6 disks attached... i see all of it as single drive with three partitions create by me... will creating tablespaces help me also is there a way i can say this table to this disk??? thx for ur help... Himanshu __

[GENERAL] Raid and tablespaces configuration

Currently my server has two raid controllers both with 5 disks attached:: On raid1: i have the database (RAID 1/0) on raid2: i have the pg_xlog (RAID 0) is this the optimal configuration or do i need to make some changes for better performance Regards Himanshu

Re: [GENERAL] Logging query plan for queries

--- Bruce Momjian wrote: > > is it possible to log plans for only the queries > whose > > duration was more than a particular value or > something > > like that... (e.g. log_min_duration_statement) > > Sure, see log_min_duration_statement. i think bruce u got my question wrong... log_min_du

[GENERAL] Logging query plan for queries

when i enable debug_print_plan and debug_pretty_print... query plans are written to log file for all queries(3 gb of logs in 3 mins) which is of course diffcult to analyze and consume too much resources is it possible to log plans for only the queries whose duration was more than a particu

Re: [GENERAL] Stats not getting updated....

Tom Lane <[EMAIL PROTECTED]> wrote: > backends only ship stats to the collector at transaction commit.> Or maybe it's at the end of processing a client command. It's certainly> not continuous.   yup that i already know but is there any way to make it do the update more frequently 4 times i

Re: [GENERAL] Stats not getting updated....

i also noted one more thing... the stats are getting updated only when i do vacuum is that wht is supposed to happen... or something is wrong   thx HimanshuHimanshu Baweja <[EMAIL PROTECTED]> wrote: i am trying to identify which tables should be moved to diff drives i first identified

[GENERAL] Stats not getting updated....

i am trying to identify which tables should be moved to diff drives i first identified the most used tables... by looking at the data in pg_statio_user_tables. and then i did sampling of the io-usage of these tables every 2 mins... to identify which are getting used when   but the probl

[GENERAL] TableSpaces across partitions and xlog

Moving table spaces along raids definately helps but will moving tablespaces across partitions help??? in my opinion it should lead to degradation of performance because read/write head will have to move more.   also, i have two raids my data is in first one... and xlog in the other. wi

Re: [GENERAL] maintenance_work_mem upper limit =1gb??

Tom Lane writes: >Himanshu Baweja <[EMAIL PROTECTED]>writes:>> why has been maintenance_work_mem and work_mem been>> restricted to 1gb... >So as not to overflow on 32-bit machines. then why not add a check during configure(before compiling) to check if its a 32 or 64 bit ma

[GENERAL] maintenance_work_mem upper limit =1gb??

why has been maintenance_work_mem and work_mem been restricted to 1gb... although i dont need it but kinda server i am working on i wont mind allocating...(32gb ram) #define MaxAllocSize ((Size) 0x3fff) /* 1 gigabyte - 1 */ also for those who dont know the max. share

[GENERAL] Bgwriter--- BufferSync() and StrategyStatsDump()

wht StrategyStatsDump prints is ARC clean buffers at LRU of T1 and T2 now lets say i have a dirty buffer at position 31st from LRU and the next one is at 3500th... in cases like this... t1_clean and t2_clean are of no use a better option would be to have a function like StrategyDirtyBuffe

Re: [GENERAL] More detailed error logging?

check out http://www.postgresql.org/docs/8.0/static/runtime-config.html   there are a couple of logging parameters which can log almost everything u want :)   Regards Himanshu__Do You Yahoo!?Tired of spam? Yahoo! Mail has the best spam protection

Re: [GENERAL] getting lock information

> >> 1) pg_locks ::: need to write trigger... and have high overhead... > > > What do you mean by "write trigger"? > > I'm sure he's imagining that he can create a trigger on pg_locks and > thereby capture lock-related events :-( well that was wht i was thinking got it now... i will recompil

Re: [GENERAL] getting lock information

thing ie they do the same quey at exactly the same time... and if i can the complete lock information... i can change the application so that they dont have to wait for each other thx Himanshu --- Neil Conway <[EMAIL PROTECTED]> wrote: > Himanshu Baweja wrote: > > is there any o

[GENERAL] getting lock information

I wanted to see which tables/transaction have acquired or are waiting for which locks as far as i know there are two ways to do it   1) pg_locks ::: need to write trigger... and have high overhead...   2) trace_locks,trace_lwlocks ... etc etc ::: well for this my server says undefined p

[GENERAL] materialized view

does postgres support materialized view.. if yes can somebody post any links which might be useful in learning how to use them... thx Himanshu__Do You Yahoo!?Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

[GENERAL] Relation between pgsql_tmp and work_mem

I read somewhere in the archives we can set work_mem to the largest size of tmp file created in the pgsql_tmp folder but how can i make those files in pgsql_tmp not be deleted after use... so that i can get the file listing for all files created during the run... currently i am doing "ls -l " s

[GENERAL] bg-writer queries regarding ( [Testperf-general] BufferSync and bgwriter )

browsing through the archives i came across this:: http://pgfoundry.org/pipermail/testperf-general/2004-December/80.html   which seems to be of the time when bgwriter was in development stage   i need some clarifications..   is there a separate list of dirty buffers and clean buffers...

Re: [GENERAL] how to calculate checkpoint_segments

got it i think i need to sleep i forgot each segment is 16 Mb... sorry for buggung u all himanshuTom Lane <[EMAIL PROTECTED]> wrote: Himanshu Baweja <[EMAIL PROTECTED]>writes:> i am trying to optimise postgres 8 running on a system is there any way to know how wh

Re: [GENERAL] how to calculate checkpoint_segments

i think there is a bug in the checkpoint warning system. i had set => checkpoint_timeout  =  2000 checkpoint_warning =  2100 checkpoint_segments = 256 (for 1st run) and 64 (for 2nd run)   now i ran my test application my Wal-logs in PGDATA/pg_xlog increased by around 1.9 GB while in my log

[GENERAL] Monitoring locks

I wanted to see which tables/transaction have acquired or are waiting for which locks as far as i know there are two ways to do it   1) pg_locks ::: but we need to continously keep getting data from the table and it also locks lock manager data structures which is not a good thing to do...

[GENERAL] how to calculate checkpoint_segments

i am trying to optimise postgres 8 running on a system is there any way to know how wht should be the value of checkpoint_segments...   i always keep fync = false since my server is fully reliable   increasing checkpoint_segments degrade the performace while checkpointing as it will have a

[GENERAL] Wht all database profiler / performance monitoring tools are available

I have been using system calls like top.. free... vmstat   Are there any better ways by which i monitor the database to get the most out of it in terms of performance Do you Yahoo!? Yahoo! Small Business - Try our new resources site!