Re: Postgres slave not catching up (on 9.2)

Поиск
Список
Период
Сортировка
От Ruben Domingo Gaspar Aparicio
Тема Re: Postgres slave not catching up (on 9.2)
Дата
Msg-id 6FCBD86BDC789C43AC696969EC34022801094212C0@CERNXCHG44.cern.ch
обсуждение исходный текст
Ответ на Postgres slave not catching up (on 9.2)  (Ruben Domingo Gaspar Aparicio <Ruben.Gaspar.Aparicio@cern.ch>)
Список pgsql-performance
> dbnasg403-12a:/vol/dodpupdbtst03 on /ORA/dbs03/PUPDBTST type nfs (rw,
> actimeo=0,hard,nointr,rsize=65536,wsize=65536,tcp,timeo=600)
> dbnasg401-12a:/vol/dodpupdbtst02 on /ORA/dbs02/PUPDBTST type nfs (rw,
> actimeo=0,hard,nointr,rsize=65536,wsize=65536,tcp,timeo=600)
> 
> to
> 
> dbnasg403-12a:/vol/dodpupdbtst03 on /ORA/dbs03/PUPDBTST type nfs
> (rw,noatime,hard,nointr,rsize=65536,wsize=65536,tcp,timeo=600)
> dbnasg401-12a:/vol/dodpupdbtst02 on /ORA/dbs02/PUPDBTST type nfs
> (rw,noatime,hard,nointr,rsize=65536,wsize=65536,tcp,timeo=600)
> 
> it did have a big impact. The profile of the recovery process on terms of calls
> changed quite a lot:
> 
> From:
> 
> [postgres@itrac1202 tmp]$  strace -p 9596 -c Process 9596 attached -
> interrupt to quit Process 9596 detached
> % time     seconds  usecs/call     calls    errors syscall
> ------ ----------- ----------- --------- --------- ----------------
> 78.73    0.217824           0    456855    381376 read
> 17.87    0.049453           0    515320           lseek
>   2.89    0.007989          12       669       669 poll
>   0.33    0.000912          25        36           open
>   0.07    0.000206           0       994       994 stat
>   0.05    0.000151           0       995       787 rt_sigreturn
>   0.05    0.000133           0       673           write
>   0.00    0.000000           0        36           close
>   0.00    0.000000           0        52           kill
> ------ ----------- ----------- --------- --------- ----------------
> 100.00    0.276668                975630    383826 total
> 

This one should read:

[root@itrac1202 ~]# strace -c -p 28073
Process 28073 attached - interrupt to quit

Process 28073 detached
% time     seconds  usecs/call     calls    errors syscall
------ ----------- ----------- --------- --------- ----------------
59.16   10.756007           5   2201974   1202832 read
40.69    7.398247           3   2367885           lseek
  0.14    0.025970         154       169           open
  0.00    0.000057           0       169           close
  0.00    0.000038           0       169           kill
  0.00    0.000033           1        29           write
  0.00    0.000000           0         1           semop
------ ----------- ----------- --------- --------- ----------------
100.00   18.180352               4570396   1202832 total


Apologies for the confusion.

В списке pgsql-performance по дате отправления:

Предыдущее
От: "Graeme B. Bell"
Дата:
Сообщение: Re: Tuning the configuration
Следующее
От: "Graeme B. Bell"
Дата:
Сообщение: Re: 8xIntel S3500 SSD in RAID10 on Dell H710p