All of lore.kernel.org
 help / color / mirror / Atom feed
* FIO hangs and shows HUGE eta while replaying blkparse.bin files - 2.1.6.1
@ 2014-04-03 15:49 Lakshmi
  2014-04-03 17:17 ` Jens Axboe
  0 siblings, 1 reply; 3+ messages in thread
From: Lakshmi @ 2014-04-03 15:49 UTC (permalink / raw)
  To: fio, Jens Axboe

[-- Attachment #1: Type: text/plain, Size: 789 bytes --]

Hello,

  I am having a strange problem. I am trying to replay the blkparsed
binary files through FIO and FIO hangs staying at a certain % and the
eta time just keeps increasing.
=====================================
job2: (g=0): rw=read, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=1
fio-2.1.6.1
Starting 1 process
Jobs: 1 (f=1): [W] [45.9% done] [0KB/0KB/0KB /s] [0/0/0 iops] [eta 00m:20s]

=====================================

Input file for FIO job to replay- file_65208400.bin

Attached are the file_65208400.bin file used in FIO and the text file
of the blktrace that was run in parallel when the above job was
running.

The trace text file doesn't show any information on what went wrong ,
and can anyone give me some pointers on debugging this?

Appreciate it.

Thanks,
Lakshmi

[-- Attachment #2: bp_FIO_Job.txt --]
[-- Type: text/plain, Size: 4421 bytes --]

  8,0   24        1     0.000000000 13301  Q  WS 17181928056 + 8 [fio]
  8,0   24        2     0.000003061 13301  G  WS 17181928056 + 8 [fio]
  8,0   24        3     0.000003877 13301  P   N [fio]
  8,0   24        4     0.000005575 13301  I  WS 17181928056 + 8 [fio]
  8,0   24        5     0.000006469 13301  U   N [fio] 1
  8,0   24        6     0.000007258 13301  D  WS 17181928056 + 8 [fio]
  8,0   24        7     0.000064844     0  C  WS 17181928056 + 8 [0]
  8,0   24        8     0.004845282 13301  Q  WS 17181928056 + 8 [fio]
  8,0   24        9     0.004846067 13301  G  WS 17181928056 + 8 [fio]
  8,0   24       10     0.004846313 13301  P   N [fio]
  8,0   24       11     0.004846844 13301  I  WS 17181928056 + 8 [fio]
  8,0   24       12     0.004847081 13301  U   N [fio] 1
  8,0   24       13     0.004847376 13301  D  WS 17181928056 + 8 [fio]
  8,0   24       14     0.004892438     0  C  WS 17181928056 + 8 [0]
  8,0   24       15     0.004902139 13301  Q  WS 17181929424 + 8 [fio]
  8,0   24       16     0.004902756 13301  G  WS 17181929424 + 8 [fio]
  8,0   24       17     0.004902995 13301  P   N [fio]
  8,0   24       18     0.004903417 13301  I  WS 17181929424 + 8 [fio]
  8,0   24       19     0.004903616 13301  U   N [fio] 1
  8,0   24       20     0.004903907 13301  D  WS 17181929424 + 8 [fio]
  8,0   24       21     0.004940526     0  C  WS 17181929424 + 8 [0]
  8,0   24       22     0.006054688 13301  Q  WS 17181929424 + 8 [fio]
  8,0   24       23     0.006055313 13301  G  WS 17181929424 + 8 [fio]
  8,0   24       24     0.006055503 13301  P   N [fio]
  8,0   24       25     0.006057286 13301  I  WS 17181929424 + 8 [fio]
  8,0   24       26     0.006057532 13301  U   N [fio] 1
  8,0   24       27     0.006057774 13301  D  WS 17181929424 + 8 [fio]
  8,0   24       28     0.006101116     0  C  WS 17181929424 + 8 [0]
  8,0   24       29     0.006111188 13301  Q  WS 17181929496 + 8 [fio]
  8,0   24       30     0.006111657 13301  G  WS 17181929496 + 8 [fio]
  8,0   24       31     0.006111838 13301  P   N [fio]
  8,0   24       32     0.006112208 13301  I  WS 17181929496 + 8 [fio]
  8,0   24       33     0.006112407 13301  U   N [fio] 1
  8,0   24       34     0.006112626 13301  D  WS 17181929496 + 8 [fio]
  8,0   24       35     0.006149003     0  C  WS 17181929496 + 8 [0]
  8,0   24       36     0.007519532 13301  Q  WS 17181929496 + 8 [fio]
  8,0   24       37     0.007520094 13301  G  WS 17181929496 + 8 [fio]
  8,0   24       38     0.007520309 13301  P   N [fio]
  8,0   24       39     0.007521969 13301  I  WS 17181929496 + 8 [fio]
  8,0   24       40     0.007522178 13301  U   N [fio] 1
  8,0   24       41     0.007523764 13301  D  WS 17181929496 + 8 [fio]
  8,0   24       42     0.007567032     0  C  WS 17181929496 + 8 [0]
  8,0   24       43     0.007577223 13301  Q  WS 17181930224 + 8 [fio]
  8,0   24       44     0.007577690 13301  G  WS 17181930224 + 8 [fio]
  8,0   24       45     0.007577885 13301  P   N [fio]
  8,0   24       46     0.007578344 13301  I  WS 17181930224 + 8 [fio]
  8,0   24       47     0.007578532 13301  U   N [fio] 1
  8,0   24       48     0.007578821 13301  D  WS 17181930224 + 8 [fio]
  8,0   24       49     0.007615989     0  C  WS 17181930224 + 8 [0]
  8,0   24       50     0.008756563 13301  Q  WS 17181930224 + 8 [fio]
  8,0   24       51     0.008757174 13301  G  WS 17181930224 + 8 [fio]
  8,0   24       52     0.008757377 13301  P   N [fio]
  8,0   24       53     0.008757782 13301  I  WS 17181930224 + 8 [fio]
  8,0   24       54     0.008757969 13301  U   N [fio] 1
  8,0   24       55     0.008758205 13301  D  WS 17181930224 + 8 [fio]
  8,0   24       56     0.008799750     0  C  WS 17181930224 + 8 [0]
CPU24 (./4K_RR_Trace):
 Reads Queued:           0,        0KiB	 Writes Queued:           8,       32KiB
 Read Dispatches:        0,        0KiB	 Write Dispatches:        8,       32KiB
 Reads Requeued:         0		 Writes Requeued:         0
 Reads Completed:        0,        0KiB	 Writes Completed:        8,       32KiB
 Read Merges:            0,        0KiB	 Write Merges:            0,        0KiB
 Read depth:             0        	 Write depth:             1
 IO unplugs:             8        	 Timer unplugs:           0

Throughput (R/W): 0KiB/s / 4,000KiB/s
Events (./4K_RR_Trace): 56 entries
Skips: 0 forward (0 -   0.0%)

[-- Attachment #3: file_65208400.bin --]
[-- Type: application/octet-stream, Size: 432 bytes --]

^ permalink raw reply	[flat|nested] 3+ messages in thread
* Re: FIO hangs and shows HUGE eta while replaying blkparse.bin files - 2.1.6.1
  2014-04-03 15:49 FIO hangs and shows HUGE eta while replaying blkparse.bin files - 2.1.6.1 Lakshmi
@ 2014-04-03 17:17 ` Jens Axboe
  2014-04-07 17:16   ` Lakshmi
  0 siblings, 1 reply; 3+ messages in thread
From: Jens Axboe @ 2014-04-03 17:17 UTC (permalink / raw)
  To: Lakshmi, fio

On 04/03/2014 09:49 AM, Lakshmi wrote:
> Hello,
>
>    I am having a strange problem. I am trying to replay the blkparsed
> binary files through FIO and FIO hangs staying at a certain % and the
> eta time just keeps increasing.
> =====================================
> job2: (g=0): rw=read, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=1
> fio-2.1.6.1
> Starting 1 process
> Jobs: 1 (f=1): [W] [45.9% done] [0KB/0KB/0KB /s] [0/0/0 iops] [eta 00m:20s]
>
> =====================================
>
> Input file for FIO job to replay- file_65208400.bin
>
> Attached are the file_65208400.bin file used in FIO and the text file
> of the blktrace that was run in parallel when the above job was
> running.
>
> The trace text file doesn't show any information on what went wrong ,
> and can anyone give me some pointers on debugging this?

If you look at the blktrace, there is a lot of time between events. Does 
it complete if you add replay_no_stall=1 as an option?

-- 
Jens Axboe



^ permalink raw reply	[flat|nested] 3+ messages in thread
* Re: FIO hangs and shows HUGE eta while replaying blkparse.bin files - 2.1.6.1
  2014-04-03 17:17 ` Jens Axboe
@ 2014-04-07 17:16   ` Lakshmi
  0 siblings, 0 replies; 3+ messages in thread
From: Lakshmi @ 2014-04-07 17:16 UTC (permalink / raw)
  To: Jens Axboe; +Cc: fio

Thanks Jens, yes that worked.

-Lakshmi

On Thu, Apr 3, 2014 at 12:17 PM, Jens Axboe <axboe@kernel.dk> wrote:
> On 04/03/2014 09:49 AM, Lakshmi wrote:
>>
>> Hello,
>>
>>    I am having a strange problem. I am trying to replay the blkparsed
>> binary files through FIO and FIO hangs staying at a certain % and the
>> eta time just keeps increasing.
>> =====================================
>> job2: (g=0): rw=read, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=1
>> fio-2.1.6.1
>> Starting 1 process
>> Jobs: 1 (f=1): [W] [45.9% done] [0KB/0KB/0KB /s] [0/0/0 iops] [eta
>> 00m:20s]
>>
>> =====================================
>>
>> Input file for FIO job to replay- file_65208400.bin
>>
>> Attached are the file_65208400.bin file used in FIO and the text file
>> of the blktrace that was run in parallel when the above job was
>> running.
>>
>> The trace text file doesn't show any information on what went wrong ,
>> and can anyone give me some pointers on debugging this?
>
>
> If you look at the blktrace, there is a lot of time between events. Does it
> complete if you add replay_no_stall=1 as an option?
>
> --
> Jens Axboe
>


^ permalink raw reply	[flat|nested] 3+ messages in thread
end of thread, other threads:[~2014-04-07 17:16 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2014-04-03 15:49 FIO hangs and shows HUGE eta while replaying blkparse.bin files - 2.1.6.1 Lakshmi
2014-04-03 17:17 ` Jens Axboe
2014-04-07 17:16   ` Lakshmi

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.