Discussions
Categories
Groups
Community Home
Categories
INTERNAL ENABLEMENT
POPULAR
PUBLIC CLOUD
PRIVATE CLOUD
Quick Links
MY LINKS
HELPFUL TIPS
Back to website
Home
Web CMS (TeamSite)
OD Receiver service stopping on rollback
System
we're getting very consistent crashes/stops of the OD service on the OD Receiver, under the conditions that the DNR fails and triggers a rollback of the transactional deployment...
OD Receiver => Version: 5.5.1 Build 11146 REL Interwoven 20020829 (Windows 2000 server)
OD Base => Version: 5.6.0.0.0 Build 15978 Interwoven 20030121 (Windows 2000 server)
the deployment section from the OD config file:
<deployment transactional="yes" >
<execDeploymentTask useDefinition="agent.com" >
<deployNRun>
<dnrDeployment location="source" when="after" state="success">
<script cmd="$dnrCommand^" async="no" />
</dnrDeployment>
</deployNRun>
</execDeploymentTask>
<execDeploymentTask useDefinition="www.com" />
</deployment>
the logs on the source side are complete, through rollback...the logs on the receiver side are incomplete, they cutoff/die right at the start of rollback, when the service stops...I can copy what little there is to here if it'll be helpful...
probably a version compatibility issue? but we can't upgrade the receivers to 5.6 quite yet, though this shouldn't be the problem...by any chance is this a known bug?
any help appreciated...
thanks,
-Rori
Find more posts tagged with
Comments
Migrateduser
OD 5.6 and 5.5.1 are compatible (i.e., you can deploy from 5.6 to 5.5.1.)
I believe there were issues related to rollback and DNRs in OD 5.5.1 that were fixed in SP2. But based on your build number, it appears you've already got SP2 installed.
Check that you're not running out of disk space on the target host. Other than that, perhaps Tech Support can recreate your scenario to determine whether the problem is with 5.5.1 or compatibility between 5.5.1 and 5.6.
Todd Scallan
Director of Product Management
Interwoven
t: 408-530-7167
e:
tscallan@interwoven.com
Migrateduser
I encountered a similar issue a while back and, if I remember correctly, it was due to the web server (IIS in this case) have a mapped virtual directory on one of the sub folders being deployed to.
That is:
IIS had a virtual directory of "/images" mapped to "D:\docroot\shared\images" and my deployment was setup to deploy into "D:\docroot".
This locked the "D:\docroot\shared\images" directory in a way that was not easily detected, then a transactional deployment would try to delete this directory (without an error) but then replacing the directory would fail causing a rollback which would also fail and this failure would NOT get reported to the OD Base sender.
Check for this situation and similar ones. You can test by changing from transactional to non-transactional deployments which does not delete directories each deploy. Alternatively, you can just stop IIS and try the deployment to see if it still fails.
If this is the case and you cannot change your configuration, you need to setup separate deployments for each subdirectory that is mapped to and exclude this directory from any other deployments. Alternatively, you can cause OD to stop IIS, deploy, and then restart IIS though this is usually not acceptable.
-------------------------------
E-Mail :
farnsaw@stonedoor.com