Job is being access from another user
Moderators: chulett, rschirm, roy
Job is being access from another user
Hi,
we have a problem with a job.
when I work with the job the network goes down and now it is impossible to access it.
I know that if I restart the server is possible tho unlock the job but now is impossible to do because some other jobs are running.
Does a method exist to unlock the job?
In DataStage Administrator there is the parameter "Inactivity timeout" set to "do not timeout". Is this parameter usefull to automatically unlock a job after a predefinited time?
Thanks in advance
Mario Loreti
we have a problem with a job.
when I work with the job the network goes down and now it is impossible to access it.
I know that if I restart the server is possible tho unlock the job but now is impossible to do because some other jobs are running.
Does a method exist to unlock the job?
In DataStage Administrator there is the parameter "Inactivity timeout" set to "do not timeout". Is this parameter usefull to automatically unlock a job after a predefinited time?
Thanks in advance
Mario Loreti
Hi,
If you want to effectively kill your jobs, make sure the option Enable job administration in Director (Administrator) is checked. You will have access to clean-up ressources in Director where you can get rid of any job. Otherwise, you can type in Administrator list.readu to know what are the active connections and then kill them from your system.
For your variable, Inactivity timeout, I do not know what it does and have never used it..sorry
Hope this helps
MAT
If you want to effectively kill your jobs, make sure the option Enable job administration in Director (Administrator) is checked. You will have access to clean-up ressources in Director where you can get rid of any job. Otherwise, you can type in Administrator list.readu to know what are the active connections and then kill them from your system.
For your variable, Inactivity timeout, I do not know what it does and have never used it..sorry
Hope this helps
MAT
-
- Participant
- Posts: 54607
- Joined: Wed Oct 23, 2002 10:52 pm
- Location: Sydney, Australia
- Contact:
Inactivity timeout is unrelated to this problem. It's specifically about inactivity in a connected client. In this case, the client has been abnormally disconnected from the server.
When you have a job design open for editing, you hold an exclusive lock in the database where job designs are stored (that is, the DataStage repository). A moment's thought will convince you that this is necessary. The owner of that lock is your agent process on the DataStage server. When your client process is abnormally disconnected, the agent process remains recorded as the owner of the lock; DataStage lock tables are maintained in shared memory.
Shutting down the server will clear all processes and locks, but is a bit extreme as a solution.
If you log in to the server with DataStage administrative privileges you will be able to access the LIST.READU command to show you which locks are held. The one you are looking for is a READU (read for update) lock on the job in question. Note its user number. Then you can execute the DS.TOOLS command, which brings up a menu that, among other things, allows locks to be administered; you may be able to release the lock via this mechanism.
If all else fails, exit from the DS.TOOLS menu by pressing Enter enough times, then execute the command UNLOCK USER user# ALL, where user# is the user number you noted earlier from LIST.READU.
WARNING!!
Do not release locks without being _absolutely sure_ that the lock is not in use for some valid purpose. The UNLOCK command is potentially very dangerous, and can result in lost development (or worse) if used arbitrarily.
Ray Wurlod
Education and Consulting Services
ABN 57 092 448 518
When you have a job design open for editing, you hold an exclusive lock in the database where job designs are stored (that is, the DataStage repository). A moment's thought will convince you that this is necessary. The owner of that lock is your agent process on the DataStage server. When your client process is abnormally disconnected, the agent process remains recorded as the owner of the lock; DataStage lock tables are maintained in shared memory.
Shutting down the server will clear all processes and locks, but is a bit extreme as a solution.
If you log in to the server with DataStage administrative privileges you will be able to access the LIST.READU command to show you which locks are held. The one you are looking for is a READU (read for update) lock on the job in question. Note its user number. Then you can execute the DS.TOOLS command, which brings up a menu that, among other things, allows locks to be administered; you may be able to release the lock via this mechanism.
If all else fails, exit from the DS.TOOLS menu by pressing Enter enough times, then execute the command UNLOCK USER user# ALL, where user# is the user number you noted earlier from LIST.READU.
WARNING!!
Do not release locks without being _absolutely sure_ that the lock is not in use for some valid purpose. The UNLOCK command is potentially very dangerous, and can result in lost development (or worse) if used arbitrarily.
Ray Wurlod
Education and Consulting Services
ABN 57 092 448 518
-
- Participant
- Posts: 6
- Joined: Wed May 07, 2003 5:29 am
- Location: India
- Contact:
In certain cases, the jobs get locked preventing access to the user and disabling deletion of the job from the project aswell. In such a scenario, the DataStage Server can be restarted which releases the lock on the job. However, since this process involves loss of time for all the users connected to the server, an alternate and simple procedure can be adopted in crunch situations. The sequence of steps are mentioned :
1. Login as administrator onto the DataStage Administrator.
2. Choose the current project from among the list of projects.
3. On the Command Center, execute the command:
LIST.READU ALL
The ouput in following format should be displayed:
Active Record Locks:
Device.... Inode..... Netnode Userno Lmode Pid Item-ID....................
1059298505 564510846 0 63404 66 RU 2132 AddressKEY1
587005274 365914125 0 63404 67 RL 2132 projpete&!DS.ADMIN!&
437701459 2037650065 0 63404 80 RU 2132 A55FD94A-4E02-48D5-A5F7-03EB
8A0847BC-ASCENT-JOE:2132
Here the job called "AddressKEY1" is locked, it associates with Inode 564510846.
4. To unlock the job, execute the command:
UNLOCK INODE 564510846 ALL
The following messages are displayed:
Clearing Record locks.
Clearing GROUP locks.
Clearing FILE locks.
5. Execute command LIST.READU ALL again, the job AddressKEY1 should disappear.
6. The job should be unlocked now.
However, the mentioned procedure must be followed only when it is absolutely impossible to restart the server.
Rajesh
1. Login as administrator onto the DataStage Administrator.
2. Choose the current project from among the list of projects.
3. On the Command Center, execute the command:
LIST.READU ALL
The ouput in following format should be displayed:
Active Record Locks:
Device.... Inode..... Netnode Userno Lmode Pid Item-ID....................
1059298505 564510846 0 63404 66 RU 2132 AddressKEY1
587005274 365914125 0 63404 67 RL 2132 projpete&!DS.ADMIN!&
437701459 2037650065 0 63404 80 RU 2132 A55FD94A-4E02-48D5-A5F7-03EB
8A0847BC-ASCENT-JOE:2132
Here the job called "AddressKEY1" is locked, it associates with Inode 564510846.
4. To unlock the job, execute the command:
UNLOCK INODE 564510846 ALL
The following messages are displayed:
Clearing Record locks.
Clearing GROUP locks.
Clearing FILE locks.
5. Execute command LIST.READU ALL again, the job AddressKEY1 should disappear.
6. The job should be unlocked now.
However, the mentioned procedure must be followed only when it is absolutely impossible to restart the server.
Rajesh
-
- Participant
- Posts: 54607
- Joined: Wed Oct 23, 2002 10:52 pm
- Location: Sydney, Australia
- Contact:
-
- Participant
- Posts: 74
- Joined: Sat Jul 08, 2006 4:09 am
- Location: Melbourne, Australia
- Contact:
I am using to unlock my jobs.
The message says:
Cant we unlock our own jobs? Otherwise everytime a job hangs, we need to trouble the DS Admin.
Code: Select all
"UNLOCK INODE 182641 ALL"
The message says:
Code: Select all
Insufficient privileges to perform lock maintenance.
Durga Prasad
You need to be the Administrator to UNLOCK jobs. In addition, you cannot unlock just an INODE, you also need to specify the DEVICE. Furthermore, you should unlock the user, not the file.
<a href=http://www.worldcommunitygrid.org/team/ ... TZ9H4CGVP1 target="WCGWin">
</a>
</a>