OpenDeploy Error : Could not read phase start message.

sumitk

Hi All,

I am trying to deploy from OD 6.0.2.1.0 to OD 6.1.1.

I am getting following error in the src log :

ERROR: Could not read phase start message.
LIB: 2008-11-21 07:46:32 Receiver-ERROR: À.
LIB: 2008-11-21 07:46:32 ERROR: Failed to deploy local_directory[./IST/lms2] of area[/iwmnt/default/main/projects/batchServerLeadMgmtSystem/STAGING]

When I tried to look through receiver logs, I got the following error:

ERROR: Source directory doesn't exists /teamsite/iw-home/tmp/batchServerLeadMgmtSystem/programs.iwnew
LIB: 2008-11-21 08:04:10 ERROR: Failed copying dir(/teamsite/iw-home/tmp/batchServerLeadMgmtSystem/programs.iwnew) to (/teamsite/iw-home/tmp/batchServerLeadMgmtSystem/programs)

None of my source directories are deleted. OD is deploying some files on target but not all which are in source. I am attaching my OD config, src and receiver logs for your reference. Appreciate any help provided in resolving this issue.

Thank you in advance.

Cheers,
Sumit

sumitk

Hi,

I could narrow down the problem. It seems that the deployment fails only for programs folder. I removed the programs folder from src and deployment went fine.

Thats weird behavior. Can anybody shed some light on this? Has anyone experienced this scenario before?

Cheers,
Sumit

Adam Stoller

Hi, I could narrow down the problem. It seems that the deployment fails only for programs folder. I removed the programs folder from src and deployment went fine. Thats weird behavior. Can anybody shed some light on this? Has anyone experienced this scenario before? Cheers, Sumit

Some questions which may provide pointers to further debugging and resolving this issue:

sumitk

Thanks Ghoti for your response. Here are answers to your question,

What are the permissions on the programs folder on both source and target systems? Is OD running as root or as some non-root user? Is this transactional or non-transactional? Are there any other deployments going on to the same target area concurrently? Do you have the teamsite.version information set correctly in the BaseServer's deploy.cfg file?

1. Later on I realized that the behavior is not specific to folder. I deleted programs folder from source and did deployment, the deployment again failed for some other folder with same error message. Programs folder has permissions 775 on source. Since this folder did not get deploy to target, we dont have this folder on target. The permissions set in deploy config are 755.
2. OD is running as non-root as I am running deployments through command line. But later on it will be run as root through workflow.
3. Its a transactional deployment.
4. There are multiple deployments going on concurrently to the same target as the target server is our testing environment. I have observed one more thing that, after one deployment (which fails), if I try to do another deployment, the deployment fails with error message, "could not connect to target, Exceeded all 3 retries ....". When I check the OD on target through iwodserverstatus, it says "not active". when I try to restart it, I get error message that please stop OD before starting. Then I have to stop OD and then restart on target. Is there anything wrong with the OD on target?
5. I could not find such entry in deploy.cfg on base server.

I am not sure, its the receiver OD has some problem or deploy config problem?

Adam Stoller

Thanks Ghoti for your response. Here are answers to your question, 1. Later on I realized that the behavior is not specific to folder. I deleted programs folder from source and did deployment, the deployment again failed for some other folder with same error message. Programs folder has permissions 775 on source. Since this folder did not get deploy to target, we dont have this folder on target. The permissions set in deploy config are 755. 2. OD is running as non-root as I am running deployments through command line. But later on it will be run as root through workflow. 3. Its a transactional deployment. 4. There are multiple deployments going on concurrently to the same target as the target server is our testing environment. I have observed one more thing that, after one deployment (which fails), if I try to do another deployment, the deployment fails with error message, "could not connect to target, Exceeded all 3 retries ....". When I check the OD on target through iwodserverstatus, it says "not active". when I try to restart it, I get error message that please stop OD before starting. Then I have to stop OD and then restart on target. Is there anything wrong with the OD on target? 5. I could not find such entry in deploy.cfg on base server. I am not sure, its the receiver OD has some problem or deploy config problem?

It sounds like you're trying to resolve several different problems all at once - I recommend trying to simplify things and see if the problems are reproducible one-by-one.

FYI - if you did not run iwodnonroot - then your deployments are run as root (Unix, or SYSTEM or Administrator on Windows) regardless of who initiates the deployment from the command line.

Concurrent transactional deployments to the same target area will not work reliably - I don't recall what version it was added for, but you basically need to serialize the deployments

Kevin Stewart

Morning all. Experiencing this exact error in v20.2 ODR.

"LIB: 2023-05-09 16:47:34  ERROR:  Could not read phase start message.
LIB: 2023-05-09 16:47:34  Receiver-ERROR: 0ï0+."

Did anyone come up with a solution or path to discovery for this?

thanks in advance

.m