[ 
https://issues.apache.org/jira/browse/SCM-142?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17960663#comment-17960663
 ] 

ASF GitHub Bot commented on SCM-142:
------------------------------------

jira-importer commented on issue #444:
URL: https://github.com/apache/maven-scm/issues/444#issuecomment-2964594930

   **[Bob 
Herrmann](https://issues.apache.org/jira/secure/[email protected])**
 commented
   
   Ok, it isnt a deleted (local) file issue.   for some reason, its not always 
getting the current contents of starteam.   Seems like it sometimes skips some 
files.   The only corrective measure I can take is to blow away the 
working-directory and let it recheck everything out.   I'm suspecting this has 
something to do with how files are checked out of starteam.
   
   stcmd co -x -nologo -stop -p bobh@starteam2:49303/PR-j/PR-j -fp 
/home/bobh/continuum-1.0.2/apps/continuum/working-directory/1 -merge 
-neverprompt -is
   
   stcmd delete-local -x -nologo -stop -p bobh@starteam2:49303/PR-j/PR-j -fp 
/home/bobh/continuum-1.0.2/apps/continuum/working-directory/1 -is -filter N
   
   These commmands look fine.  Although we have had problems doing incremental 
builds with AntHill too.  So I suspect something funny with starteam.
   




> Starteam tree stale
> -------------------
>
>                 Key: SCM-142
>                 URL: https://issues.apache.org/jira/browse/SCM-142
>             Project: Maven SCM (Moved to GitHub Issues)
>          Issue Type: Bug
>          Components: maven-scm-provider-starteam
>         Environment: continuum-1.0.2
>            Reporter: Bob Herrmann
>            Priority: Major
>
> It only takes a few changes to starteam for the checked out filesystem to 
> become hopelessly stale.  The only recovery option is to completely remove 
> all files and startover.  Either the code checking out is not handing the 
> timestamps correctly, or starteam command line has problems keeping a checked 
> out tree in sync (this more likely - as we also see this problem with AntHill 
> when using it's incremental builder.)   Possible fixes might be to detect 
> 'unknown' status's and flush the checked out tree.  Or try using the 2005 bco 
> command instead of stcmd.  



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

Reply via email to