It's all about the answers!

Ask a question

Getting "Failure Downloading file"


Kevin Ramer (4.5k8183200) | asked Aug 23 '11, 9:33 p.m.
Seeing this kind of error using Ant build toolkit

Not using a proxy to reach https://rtp-rtc3:9443/jazz
Fetching contents of workspace "ITCAM_WR_720ProductionBuildWorkspace2" to "/ausshr2/itcam_wr_720_rtc/prod/TOP/src".
com.ibm.team.repository.common.TeamRepositoryException: Failure downloading file /itcam.installer/ecam/src/itcamfwas/data/graphics/ko/itcamfwas_dc_13.gif
at com.ibm.team.filesystem.client.internal.load.MergeLoadMutator.storeFileContents(MergeLoadMutator.java:1432)
at com.ibm.team.filesystem.client.internal.load.MergeLoadMutator.access$0(MergeLoadMutator.java:1411)
at com.ibm.team.filesystem.client.internal.load.MergeLoadMutator$DownloadHandler.downloadStreamAcquired(MergeLoadMutator.java:143)
at com.ibm.team.filesystem.client.FileDownloadHandler.downloadStreamAcquired(FileDownloadHandler.java:86)
at com.ibm.team.filesystem.client.FileDownloadHandler.downloadStreamAcquired(FileDownloadHandler.java:1)

....

teamFetch] Caused by: java.io.FileNotFoundException: /ausshr2/itcam_wr_720_rtc/mtpt1/TOP/src/itcam.installer/ecam/src/itcamfwas/data/graphics/ko/itcamfwas_dc_13.gif (A system call received an interrupt.)
at java.io.FileOutputStream.open(Native Method)
at java.io.FileOutputStream.<init>(FileOutputStream.java:205)
at java.io.FileOutputStream.<init>(FileOutputStream.java:157)
at com.ibm.team.filesystem.client.internal.load.MergeLoadMutator.internalStoreFileContents(MergeLoadMutator.java:1448)
... 11 more
com.ibm.team.build.internal.scm.SourceControlUtility$2: Status ERROR: com.ibm.team.filesystem.client code=0 There were errors during download. It is recommended that components be reloaded. The reload will be incremental, loading only the missing items. You can further reduce the reload time by reloading just the projects out of sync. null children=


Lead-in work:
Location is prepared from a tar'd image of a previous build (which includes
.jazz5) and a team accept run.
Then that tree gets mv'd and a rm -r -f on the mv'd target is put in the background then the team fetch is run.

My conjecture is that somehow the rm is somehow knocking the stilts off of the
fetch. My colleague seems to think that renaming the top level directory will
protect enough, BUT I think what if the .jazz5 meta data contains things like iNodes (which aren't changed by rename) and the team fetch is actually trying
to do its work in the old directory....

The reason for the move and remove is that deleted objects in the stream aren't
deleted during the fetch by the ant tooling.

Searches for this kind of error in the RTC work items and found few examples.

Any thoughts ?

6 answers



permanent link
Kevin Ramer (4.5k8183200) | answered Sep 06 '11, 10:21 a.m.
Seeing this kind of error using Ant build toolkit

Not using a proxy to reach https://rtp-rtc3:9443/jazz
Fetching contents of workspace "ITCAM_WR_720ProductionBuildWorkspace2" to "/ausshr2/itcam_wr_720_rtc/prod/TOP/src".
com.ibm.team.repository.common.TeamRepositoryException: Failure downloading file /itcam.installer/ecam/src/itcamfwas/data/graphics/ko/itcamfwas_dc_13.gif
at com.ibm.team.filesystem.client.internal.load.MergeLoadMutator.storeFileContents(MergeLoadMutator.java:1432)
at com.ibm.team.filesystem.client.internal.load.MergeLoadMutator.access$0(MergeLoadMutator.java:1411)
at com.ibm.team.filesystem.client.internal.load.MergeLoadMutator$DownloadHandler.downloadStreamAcquired(MergeLoadMutator.java:143)
at com.ibm.team.filesystem.client.FileDownloadHandler.downloadStreamAcquired(FileDownloadHandler.java:86)
at com.ibm.team.filesystem.client.FileDownloadHandler.downloadStreamAcquired(FileDownloadHandler.java:1)

....

teamFetch] Caused by: java.io.FileNotFoundException: /ausshr2/itcam_wr_720_rtc/mtpt1/TOP/src/itcam.installer/ecam/src/itcamfwas/data/graphics/ko/itcamfwas_dc_13.gif (A system call received an interrupt.)
at java.io.FileOutputStream.open(Native Method)
at java.io.FileOutputStream.<init>(FileOutputStream.java:205)
at java.io.FileOutputStream.<init>(FileOutputStream.java:157)
at com.ibm.team.filesystem.client.internal.load.MergeLoadMutator.internalStoreFileContents(MergeLoadMutator.java:1448)
... 11 more
com.ibm.team.build.internal.scm.SourceControlUtility$2: Status ERROR: com.ibm.team.filesystem.client code=0 There were errors during download. It is recommended that components be reloaded. The reload will be incremental, loading only the missing items. You can further reduce the reload time by reloading just the projects out of sync. null children=


Lead-in work:
Location is prepared from a tar'd image of a previous build (which includes
.jazz5) and a team accept run.
Then that tree gets mv'd and a rm -r -f on the mv'd target is put in the background then the team fetch is run.

My conjecture is that somehow the rm is somehow knocking the stilts off of the
fetch. My colleague seems to think that renaming the top level directory will
protect enough, BUT I think what if the .jazz5 meta data contains things like iNodes (which aren't changed by rename) and the team fetch is actually trying
to do its work in the old directory....

The reason for the move and remove is that deleted objects in the stream aren't
deleted during the fetch by the ant tooling.

Searches for this kind of error in the RTC work items and found few examples.

Any thoughts ?
\


Anyone ?

permanent link
Tim Mok (6.6k38) | answered Sep 06 '11, 11:24 a.m.
JAZZ DEVELOPER
com.ibm.team.build.internal.scm.SourceControlUtility$2: Status ERROR: com.ibm.team.filesystem.client code=0 There were errors during download. It is recommended that components be reloaded. The reload will be incremental, loading only the missing items. You can further reduce the reload time by reloading just the projects out of sync. null children=
It looks like the workspace is out of sync. I assume the tar of the previous build is the build's workspace. If you've run builds after that tar was made, the build workspace may accept changes. This makes the metadata for the build workspace in the tar out of sync with what is actually on the server.

permanent link
Kevin Ramer (4.5k8183200) | answered Sep 30 '11, 10:30 a.m.
This is continuing to occur on about 1/2 of the build attempts. In tracking down
the affected file (not always the same one !) the change history more often than
not has only "Share Projects" as its sole change.

There are inconsistencies when comparing the build workspace with the stream. These are mostly outbound changes which appear to be component baselines.


Outgoing Changes
Component _Build_root
Component automation-test
Component build
JAKUB A. PIECHNIK <jakub> 12239: Fixing break build - Fixing build break 2011/09/29
Component itcam.adtip
Component itcam.ant
Component itcam.common
Component itcam.configurator
Component itcam.dc
Marvin Coley <mecoley> 11721: AD0000C AD72 zOS unconfig.sh remove plugin 2011/09/29
Component itcam.de
Component itcam.demo
Component itcam.installer
Component itcam.ms
Component itcam.msdc
Component itcam.tcr
Component itcam.tema
Component itcam.toolkit
Incoming Changes
Component _Build_root
Component automation-test
Component build
Component itcam.adtip
Component itcam.ant
Component itcam.common
Component itcam.configurator
Component itcam.dc
Component itcam.de
Component itcam.demo
Component itcam.installer
Component itcam.ms
Component itcam.msdc
Component itcam.tcr
Component itcam.tema
Component itcam.toolkit


This particular stream has had some unintended component replacements along the way. In fact I caused one, but I was trying to compare two baselines !!!! I almost came out of my chair when I saw the RTC client put "Replacing component baseline" in the status area

If I could figure out how to put an image in this post I'd show the Pending Changes view of the build workspace. I see "img" tag but I've nowhere to post.

Anyone have any suggestions (about the "File not found" not the img ) :-)


com.ibm.team.build.internal.scm.SourceControlUtility$2: Status ERROR: com.ibm.team.filesystem.client code=0 There were errors during download. It is recommended that components be reloaded. The reload will be incremental, loading only the missing items. You can further reduce the reload time by reloading just the projects out of sync. null children=
It looks like the workspace is out of sync. I assume the tar of the previous build is the build's workspace. If you've run builds after that tar was made, the build workspace may accept changes. This makes the metadata for the build workspace in the tar out of sync with what is actually on the server.

permanent link
Tim Mok (6.6k38) | answered Sep 30 '11, 10:47 a.m.
JAZZ DEVELOPER
Are you still doing the same thing with the build and keeping the .jazz5 directory?

This is what causes your out-of-sync issue. The build workspace has changed since the last time the data was fetched. The .jazz5 has metadata that confuses the fetch operation.

permanent link
Kevin Ramer (4.5k8183200) | answered Oct 04 '11, 7:12 p.m.
Well,
I think this gets done for all the RTC based builds we run, but this particular build is the only one of probably 100 or more RTC source builds that exhibit this failure. If it were the .jazz5, I'd think that there would be some percentage (randomly) that would fail in a similar fashion.

Are you still doing the same thing with the build and keeping the .jazz5 directory?

This is what causes your out-of-sync issue. The build workspace has changed since the last time the data was fetched. The .jazz5 has metadata that confuses the fetch operation.



permanent link
Kevin Ramer (4.5k8183200) | answered Nov 02 '11, 10:49 a.m.
Are you still doing the same thing with the build and keeping the .jazz5 directory?

This is what causes your out-of-sync issue. The build workspace has changed since the last time the data was fetched. The .jazz5 has metadata that confuses the fetch operation.



The 'fetch' is being done into an empty location. The workspace changes are accepted as step 1 of the RTC functions. I cannot find any information about this message:

Caused by: java.io.FileNotFoundException: .../docs/manual/api/org/apache/tools/mail/SmtpResponseReader.html (A system call received an interrupt.)

I'm thinking that the file was created and possibly the write or some other system call found something amiss. This is being done into NFS file system.

Your answer


Register or to post your answer.


Dashboards and work items are no longer publicly available, so some links may be invalid. We now provide similar information through other means. Learn more here.