Mailing-List: contact whirr-dev-help@incubator.apache.org; run by ezmlm
Precedence: bulk
Reply-To: whirr-dev@incubator.apache.org
Date: Fri, 29 Jul 2011 10:01:09 +0000 (UTC)
From: "Adrian Cole (JIRA)" <jira@apache.org>
To: whirr-dev@incubator.apache.org
Message-ID: 
 <432677347.18200.1311933669800.JavaMail.tomcat@hel.zones.apache.org>
In-Reply-To: 
 <791965039.18083.1311926770340.JavaMail.tomcat@hel.zones.apache.org>
Subject: [jira] [Commented] (WHIRR-349) Retry if blobstore put fails
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/WHIRR-349?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13072760#comment-13072760 ] 

Adrian Cole commented on WHIRR-349:
-----------------------------------

the code in BlobCache needs to be revised.  jclouds will by default retry any replayable payloads.  InputStreams are not replayable.  Looks like the smoking gun is here where a file is converted to a stream:


  public synchronized void putIfAbsent(File file) throws FileNotFoundException {
    putIfAbsent(file.getName(), new FileInputStream(file), file.length());
  }

It is better to send the File, byte[], String, etc. directly than do this.  All of those payloads are retryable.  I hope this helps.

P.S. we should use BlobBuilder in BlobCache as it looks nicer

> Retry if blobstore put fails
> ----------------------------
>
>                 Key: WHIRR-349
>                 URL: https://issues.apache.org/jira/browse/WHIRR-349
>             Project: Whirr
>          Issue Type: Improvement
>          Components: core
>    Affects Versions: 0.6.0
>         Environment: Ubuntu 11.04
> java version "1.6.0_26"
> Java(TM) SE Runtime Environment (build 1.6.0_26-b03)
> Java HotSpot(TM) 64-Bit Server VM (build 20.1-b02, mixed mode)
> (using https://github.com/bdumon/whirr 13cd70e which uses a jclouds 1.1.0-SNAPSHOT)
>            Reporter: Karel Vervaeke
>            Priority: Minor
>   Original Estimate: 4h
>  Remaining Estimate: 4h
>
> When the blobstore fails to upload a file, it might be worth retrying a couple of times rather than just aborting
> the whole process.
> OTOH, the failure would occur *before* any machines are launched, so the loss is limited.
> Here's an example stacktrace of a failed upload:
> Created blob cache container 'fr6mxqa4d6aa' located in '[id=EU, scope=REGION, description=EU, parent=aws-s3, iso3166Codes=[IE], metadata={}]'
> Uploading 'mydecorator-1.0-SNAPSHOT.tar.gz' to 'fr6mxqa4d6aa' blob cache.
> Jul 29, 2011 8:53:50 AM org.jclouds.logging.jdk.JDKLogger logWarn
> WARNING: Cannot retry after server error, command is not replayable: [method=AWSS3AsyncClient.putObject, request=PUT https://fr6mxqa4d6aa.s3.amazonaws.com/mydecorator-1.0-SNAPSHOT.tar.gz HTTP/1.1]
> Exception in thread "main" org.jclouds.http.HttpResponseException: error after writing 335872/637480 bytes to PUT https://fr6mxqa4d6aa.s3.amazonaws.com/mydecorator-1.0-SNAPSHOT.tar.gz HTTP/1.1 connecting to PUT https://fr6mxqa4d6aa.s3.amazonaws.com/mydecorator-1.0-SNAPSHOT.tar.gz HTTP/1.1
> 	at org.jclouds.http.internal.BaseHttpCommandExecutorService$HttpResponseCallable.call(BaseHttpCommandExecutorService.java:183)
> 	at org.jclouds.http.internal.BaseHttpCommandExecutorService$HttpResponseCallable.call(BaseHttpCommandExecutorService.java:134)
> 	at java.util.concurrent.FutureTask$Sync.innerRun(FutureTask.java:334)
> 	at java.util.concurrent.FutureTask.run(FutureTask.java:166)
> 	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1110)
> 	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:603)
> 	at java.lang.Thread.run(Thread.java:679)
> Caused by: java.lang.RuntimeException: error after writing 335872/637480 bytes to PUT https://fr6mxqa4d6aa.s3.amazonaws.com/mydecorator-1.0-SNAPSHOT.tar.gz HTTP/1.1
> 	at org.jclouds.http.internal.JavaUrlHttpCommandExecutorService.convert(JavaUrlHttpCommandExecutorService.java:245)
> 	at org.jclouds.http.internal.JavaUrlHttpCommandExecutorService.convert(JavaUrlHttpCommandExecutorService.java:77)
> 	at org.jclouds.http.internal.BaseHttpCommandExecutorService$HttpResponseCallable.call(BaseHttpCommandExecutorService.java:156)
> 	... 6 more
> Caused by: java.io.IOException: Error writing request body to server
> 	at sun.net.www.protocol.http.HttpURLConnection$StreamingOutputStream.checkError(HttpURLConnection.java:2716)
> 	at sun.net.www.protocol.http.HttpURLConnection$StreamingOutputStream.write(HttpURLConnection.java:2699)
> 	at com.google.common.io.CountingOutputStream.write(CountingOutputStream.java:51)
> 	at com.google.common.io.ByteStreams.copy(ByteStreams.java:172)
> 	at org.jclouds.io.payloads.BasePayload.writeTo(BasePayload.java:67)
> 	at org.jclouds.http.internal.JavaUrlHttpCommandExecutorService.convert(JavaUrlHttpCommandExecutorService.java:243)
> 	... 8 more
> Removing blob cache 'fr6mxqa4d6aa'

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira