Could not read next directive error

Hi All -

We are facing a strange issue in one of our OD deployment scripts. This script throws an intermittent error. When I execute the script it doesn't always throws the error but sometimes.. even if I follow the same series of steps with same file list. There are some special characters recorded as well.

Error is:
LIB: 2011-04-21 03:02:00 COMMITTING Transactional Deployment
LIB 00020: 2011-04-21 03:02:10 ERROR: Could not read next directive.
LIB: 2011-04-21 03:02:10 ERROR: Failed to deploy local_directory[.] of area[/iwmnt/hip/main/us/STAGING] for filelist
-----------
LIB: 2011-04-21 03:02:26 Receiver-ERROR: Â¿)Ã^LÃ)Ã.
LIB: 2011-04-21 03:02:26 ERROR: Failed to deploy local_directory[.] of area[/iwmnt/hip/main/us/pearsonhighered/STAGING] for filelist

This deployment script makes call to a DNR as well. Could it be that the DNR is causing some issues, example, timeout?

Please suggest.

Thanks!
Payal

Env: TS/ OD 7.1 (RHEL 5.5)

Find more posts tagged with

Comments

Rick Poulin

Problem #1 is that you've installed OpenDeploy on a platform that doesn't support it. OD7.1 supports up to RHEL 5.3 only. I think OD 7.2.1 supports RHEL 5.x so you might want to consider either upgrading OD or downgrading your OS.

"Could not read next directive" probably means that the data stream was interrupted early. I'd suggest checking firewalls, intrusive packet inspectors, over-eager virus scans, disk space on both the base and receiver, that sort of thing. It would also be useful if you could post your deployment config to see if you're doing anything weird.

PayalGarg

Hi There!!

Thanks for your reply. I just checked OD release notes and found that the OS version information that you provided is indeed correct. Smiley Happy

I've attached the configuration file as well.

Thanks!
Payal

Rick Poulin

Couple thoughts..
- the special characters are probably caused by reading the log file as an ISO-8859 encoding when it looks like UTF-8.
- if you're concerned about your DNR taking too long, you might want to try commenting it out and then firing off a whole bunch of deployments to see if the problem happens again (in DEV). If so, maybe change your timeout values on the localNode tag and/or odbase/odrcvr configs. Personally, I haven't seen a need to specify those attributes on the localNode tag in the deployment config -- the odbase/odrcvr globals are usually enough. I can't see anything unusual in your config otherwise.
- I don't suspect your DNR is the problem though, from the looks of it. I'm still thinking permissions or something, because your problem is happening in the COMMIT phase, which is before your DNR is set to execute.

I'm thinking there may be something interesting in the receiver logs (of whichever one of your three receivers caused the failure), since your log snippet looks to be from the base server.

PayalGarg

Hi Rick,

From reciever end, logs are:

ENG: Failed Commit false
LIB: id=m1745 2011-04-01 14:29:04 FailedCommit 0.
LIB: (3) id=m1745 2011-04-01 14:29:04 FailedCommitFlag = 0
LIB: (3) id=m1745 2011-04-01 14:29:04 FailedCommitFromEngine = 0
LIB: id=m1745 2011-04-01 14:29:04
Directories deployed : 0 Files deployed : 1 Links deployed : 0
Directories failed : 0 Files failed : 0 Links failed : 0
Directories deleted : 0 Files deleted : 0 Links deleted : 0
LIB: 2011-04-01 14:29:04 Sender-ERROR: .
LIB: 2011-04-01 14:29:04 ERROR: Failed deploying filelist[0]
LIB: 2011-04-01 14:29:04 ERROR: Failed 1 of 1 filelists
ENG: Thread-368646 After COMMIT1 res=-1
ENG: Thread-368646 exiting commit
ENG: Thread-368646 entering awaiting-rollback
ENG: Thread-368646 Got task failed in COMMIT
ENG: Thread-368646 still waiting for 0
ENG: Thread-368646 exiting awaiting-rollback
ENG: Thread-368646 entering rollback
ENG: Thread-368646 failedCommit=false shouldWait=true
LIB: id=m1745 2011-04-01 14:29:04 FailedCommit 0.
LIB: 2011-04-01 14:29:04 ERROR: Could not read rollback start state.
LIB 00039: 2011-04-01 14:29:04 ERROR: Unable to do hand shake with the sender
LIB: id=m1745 2011-04-01 14:29:04 ROLLBACK Transactional Deployment

The only issue that I'm trying to understand here is that this configuration is not failing always. It works fine at times and sometimes it fails, even if I use same file list to be deployed.

Thanks
Payal