Hi, All
I have a daily running job. It is a control job to call anther job. It has been in use for month. But last night, It happend a serious problem. The warning message is below:
DataStage Job 831 Phantom 926
[1077] Done : DSD.RUN FactFRRevExp 0/2000 ServerName=xxxxx UserName=yyyyy
UserPassword=LDID9 AccountPeriod=200110 StagingAccountName=prod DSJobController=FactFRRevExpControl
Self Deadlatch error on device 35414440, inode 34082, group 0, latch 80000 to 10000.
This job status remained "RUNNING". After check it, found actually it had stopped running. So I cleared its job status, re-compiled, re-ran it. It works fine.
Can anyone tell me what happened last night? How to avoid this problem happen again?
Thanks in advance
Xiong
Self Deadlatch error
Hi, Jack
Thanks for your help. This job is a control job, there is no stage within it. Only codes in Job control panel of Job Parameter call another job B processing. The source and destination table of job B are definitely in different databases.
I have many same control job like this, they all work fine. Even this job, it works normally before and current, except the night before yesterday.
I just try to make sense of what caused this problem. This job produces very important data every night, and the accident like this is inacceptable.
Regards
Xiong
> -----Original Message-----
> From: Jack Luchjenbroers [SMTP:jack@luchjenbroers.com]
> Sent: Friday, 2 November 2001 11:45
> To: datastage-users@oliver.com
> Subject: Re: Self Deadlatch error
>
> Xiong
>
> At first glance I would surgest that you are reading and writing to
> the same source and have encountered Deadlock.
>
> I have had this where the Main Source is the same as the Update. A
> simple technique I have used where I encounter this is to write update
> data to a Temporary Store (Sequential File or Hash File or UV Table).
> Then in the same job use the Temporary Store to run the update.
>
> The choices of Sequential File or Hash File or UV Table are in
> Decreasing Order of Expected Performance. Performance increases in
> relevence wrt the number of expected rows.
>
>
> Hope this helps Jack
>
>
>
>
> ----- Original Message -----
> From: "Li, Xiong (Corporate Manukau)"
> To:
> Sent: Wednesday, October 31, 2001 8:09 PM
> Subject: Self Deadlatch error
>
>
> > Hi, All
> >
> > I have a daily running job. It is a control job to call anther job.
> > It
> has
> > been in use for month. But last night, It happend a serious problem.
> > The warning message is below:
> >
> > DataStage Job 831 Phantom 926
> > [1077] Done : DSD.RUN FactFRRevExp 0/2000 ServerName=xxxxx
> UserName=yyyyy
> > UserPassword=LDID9 AccountPeriod=200110 StagingAccountName=prod
> > DSJobController=FactFRRevExpControl
> > Self Deadlatch error on device 35414440, inode 34082, group 0, latch
> 80000
> > to 10000.
> >
> > This job status remained "RUNNING". After check it, found actually
> > it
> had
> > stopped running. So I cleared its job status, re-compiled, re-ran
> > it. It works fine.
> >
> > Can anyone tell me what happened last night? How to avoid this
> > problem happen again?
> >
> > Thanks in advance
> >
> > Xiong
> >
Thanks for your help. This job is a control job, there is no stage within it. Only codes in Job control panel of Job Parameter call another job B processing. The source and destination table of job B are definitely in different databases.
I have many same control job like this, they all work fine. Even this job, it works normally before and current, except the night before yesterday.
I just try to make sense of what caused this problem. This job produces very important data every night, and the accident like this is inacceptable.
Regards
Xiong
> -----Original Message-----
> From: Jack Luchjenbroers [SMTP:jack@luchjenbroers.com]
> Sent: Friday, 2 November 2001 11:45
> To: datastage-users@oliver.com
> Subject: Re: Self Deadlatch error
>
> Xiong
>
> At first glance I would surgest that you are reading and writing to
> the same source and have encountered Deadlock.
>
> I have had this where the Main Source is the same as the Update. A
> simple technique I have used where I encounter this is to write update
> data to a Temporary Store (Sequential File or Hash File or UV Table).
> Then in the same job use the Temporary Store to run the update.
>
> The choices of Sequential File or Hash File or UV Table are in
> Decreasing Order of Expected Performance. Performance increases in
> relevence wrt the number of expected rows.
>
>
> Hope this helps Jack
>
>
>
>
> ----- Original Message -----
> From: "Li, Xiong (Corporate Manukau)"
> To:
> Sent: Wednesday, October 31, 2001 8:09 PM
> Subject: Self Deadlatch error
>
>
> > Hi, All
> >
> > I have a daily running job. It is a control job to call anther job.
> > It
> has
> > been in use for month. But last night, It happend a serious problem.
> > The warning message is below:
> >
> > DataStage Job 831 Phantom 926
> > [1077] Done : DSD.RUN FactFRRevExp 0/2000 ServerName=xxxxx
> UserName=yyyyy
> > UserPassword=LDID9 AccountPeriod=200110 StagingAccountName=prod
> > DSJobController=FactFRRevExpControl
> > Self Deadlatch error on device 35414440, inode 34082, group 0, latch
> 80000
> > to 10000.
> >
> > This job status remained "RUNNING". After check it, found actually
> > it
> had
> > stopped running. So I cleared its job status, re-compiled, re-ran
> > it. It works fine.
> >
> > Can anyone tell me what happened last night? How to avoid this
> > problem happen again?
> >
> > Thanks in advance
> >
> > Xiong
> >
Log this with Ascential Tech Support. I put a call in
about 1 month ago for this same issue. Were on an
E10k vs 2.7 running DS 4.1. We get this once in
awhile on batch jobs, no apparent pattern. This is
something new to 4.1.
Good luck!
-Ken
--- "Li, Xiong (Corporate Manukau)"
wrote:
> Hi, All
>
> I have a daily running job. It is a control job to
> call anther job. It has
> been in use for month. But last night, It happend a
> serious problem. The
> warning message is below:
>
> DataStage Job 831 Phantom 926
> [1077] Done : DSD.RUN FactFRRevExp 0/2000
> ServerName=xxxxx UserName=yyyyy
> UserPassword=LDID9 AccountPeriod=200110 StagingAccountName=prod
> DSJobController=FactFRRevExpControl
> Self Deadlatch error on device 35414440, inode
> 34082, group 0, latch 80000
> to 10000.
>
> This job status remained "RUNNING". After check it,
> found actually it had
> stopped running. So I cleared its job status,
> re-compiled, re-ran it. It
> works fine.
>
> Can anyone tell me what happened last night? How to
> avoid this problem
> happen again?
>
> Thanks in advance
>
> Xiong
__________________________________________________
Do You Yahoo!?
Make a great connection at Yahoo! Personals. http://personals.yahoo.com
about 1 month ago for this same issue. Were on an
E10k vs 2.7 running DS 4.1. We get this once in
awhile on batch jobs, no apparent pattern. This is
something new to 4.1.
Good luck!
-Ken
--- "Li, Xiong (Corporate Manukau)"
wrote:
> Hi, All
>
> I have a daily running job. It is a control job to
> call anther job. It has
> been in use for month. But last night, It happend a
> serious problem. The
> warning message is below:
>
> DataStage Job 831 Phantom 926
> [1077] Done : DSD.RUN FactFRRevExp 0/2000
> ServerName=xxxxx UserName=yyyyy
> UserPassword=LDID9 AccountPeriod=200110 StagingAccountName=prod
> DSJobController=FactFRRevExpControl
> Self Deadlatch error on device 35414440, inode
> 34082, group 0, latch 80000
> to 10000.
>
> This job status remained "RUNNING". After check it,
> found actually it had
> stopped running. So I cleared its job status,
> re-compiled, re-ran it. It
> works fine.
>
> Can anyone tell me what happened last night? How to
> avoid this problem
> happen again?
>
> Thanks in advance
>
> Xiong
__________________________________________________
Do You Yahoo!?
Make a great connection at Yahoo! Personals. http://personals.yahoo.com
Xiong
At first glance I would surgest that you are reading and writing to the same source and have encountered Deadlock.
I have had this where the Main Source is the same as the Update. A simple technique I have used where I encounter this is to write update data to a Temporary Store (Sequential File or Hash File or UV Table). Then in the same job use the Temporary Store to run the update.
The choices of Sequential File or Hash File or UV Table are in Decreasing Order of Expected Performance. Performance increases in relevence wrt the number of expected rows.
Hope this helps Jack
----- Original Message -----
From: "Li, Xiong (Corporate Manukau)"
To:
Sent: Wednesday, October 31, 2001 8:09 PM
Subject: Self Deadlatch error
> Hi, All
>
> I have a daily running job. It is a control job to call anther job. It
> has been in use for month. But last night, It happend a serious
> problem. The warning message is below:
>
> DataStage Job 831 Phantom 926
> [1077] Done : DSD.RUN FactFRRevExp 0/2000 ServerName=xxxxx
UserName=yyyyy
> UserPassword=LDID9 AccountPeriod=200110 StagingAccountName=prod
> DSJobController=FactFRRevExpControl
> Self Deadlatch error on device 35414440, inode 34082, group 0, latch
> 80000 to 10000.
>
> This job status remained "RUNNING". After check it, found actually it
> had stopped running. So I cleared its job status, re-compiled, re-ran
> it. It works fine.
>
> Can anyone tell me what happened last night? How to avoid this problem
> happen again?
>
> Thanks in advance
>
> Xiong
>
At first glance I would surgest that you are reading and writing to the same source and have encountered Deadlock.
I have had this where the Main Source is the same as the Update. A simple technique I have used where I encounter this is to write update data to a Temporary Store (Sequential File or Hash File or UV Table). Then in the same job use the Temporary Store to run the update.
The choices of Sequential File or Hash File or UV Table are in Decreasing Order of Expected Performance. Performance increases in relevence wrt the number of expected rows.
Hope this helps Jack
----- Original Message -----
From: "Li, Xiong (Corporate Manukau)"
To:
Sent: Wednesday, October 31, 2001 8:09 PM
Subject: Self Deadlatch error
> Hi, All
>
> I have a daily running job. It is a control job to call anther job. It
> has been in use for month. But last night, It happend a serious
> problem. The warning message is below:
>
> DataStage Job 831 Phantom 926
> [1077] Done : DSD.RUN FactFRRevExp 0/2000 ServerName=xxxxx
UserName=yyyyy
> UserPassword=LDID9 AccountPeriod=200110 StagingAccountName=prod
> DSJobController=FactFRRevExpControl
> Self Deadlatch error on device 35414440, inode 34082, group 0, latch
> 80000 to 10000.
>
> This job status remained "RUNNING". After check it, found actually it
> had stopped running. So I cleared its job status, re-compiled, re-ran
> it. It works fine.
>
> Can anyone tell me what happened last night? How to avoid this problem
> happen again?
>
> Thanks in advance
>
> Xiong
>