[jira] Commented: (HADOOP-6208) Block loss in S3FS due to S3 inconsistency on file rename

["Bradley Buda (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/HADOOP-6208?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12760967#action_12760967 ] 

Bradley Buda commented on HADOOP-6208:
--------------------------------------

To answer the question about subsequent reads, no, unfortunately, getting one read of the latest version of an object does not guarantee that that object is fully propagated and that you won't get any more stale reads.  I ran a quick test yesterday to see how likely this scenario is (stale read after fresh read from the same client) and I was able to find one instance of a good read followed by a bad read over ~10,000 file writes (about 420,000 total reads).  For comparison, I found 7 instances in the same test where the very first read after a write was stale, but subsequent reads were all of the newest version.  The end result here is that with the current behavior, we'd have 7 truncated files of 10,000 written; with a patch to verify writes, we'd only get 1 truncated file out of 10,000.  Since the patch is simple and the verification step is cheap, I think it's worth the tradeoff; I'll let my script run a little longer and see if I can gather more significant data.

Tom: I think the safest thing to do is to verify writes (poll for consistency) after each INode write.  Looking through the code, this happens in 4 places: on directory creation, on file close, on file flush, and on rename.  Verifying file close and flush prevents truncated files as in this issue; verifying renames prevents a situation where a file can be renamed, then data can be written to the old filename (which should no longer exist), resulting in two copies of the file pointing to the same blocks (sort of like a weird hard link).  Verifying directory creates just prevents errors where a newly created directory doesn't show up for a little while; I don't think there are any data-loss issues possible in that scenario.

I have a patch and unit test for verifying in these places; I have a bit more testing to do, but I should be able to submit it to JIRA later today.

> Block loss in S3FS due to S3 inconsistency on file rename
> ---------------------------------------------------------
>
>                 Key: HADOOP-6208
>                 URL: https://issues.apache.org/jira/browse/HADOOP-6208
>             Project: Hadoop Common
>          Issue Type: Bug
>          Components: fs/s3
>    Affects Versions: 0.20.0
>         Environment: Ubuntu Linux 8.04 on EC2, Mac OS X 10.5, likely to affect any Hadoop environment
>            Reporter: Bradley Buda
>         Attachments: S3FSConsistencyTest.java
>
>
> Under certain S3 consistency scenarios, Hadoop's S3FileSystem can 'truncate' files, especially when writing reduce outputs.  We've noticed this at tracksimple where we use the S3FS as the direct input and output of our MapReduce jobs.  The symptom of this problem is a file in the filesystem that is an exact multiple of the FS block size - exactly 32MB, 64MB, 96MB, etc. in length.
> The issue appears to be caused by renaming a file that has recently been written, and getting a stale INode read from S3.  When a reducer is writing job output to the S3FS, the normal series of S3 key writes for a 3-block file looks something like this:
> Task Output:
> 1) Write the first block (block_99)
> 2) Write an INode (/myjob/_temporary/_attempt_200907142159_0306_r_000133_0/part-00133.gz) containing [block_99]
> 3) Write the second block (block_81)
> 4) Rewrite the INode with new contents [block_99, block_81]
> 5) Write the last block (block_-101)
> 6) Rewrite the INode with the final contents [block_99, block_81, block_-101]
> Copy Output to Final Location (ReduceTask#copyOutput):
> 1) Read the INode contents from /myjob/_temporary/_attempt_200907142159_0306_r_000133_0/part-00133.gz, which gives [block_99, block_81, block_-101]
> 2) Write the data from #1 to the final location, /myjob/part-00133.gz
> 3) Delete the old INode 
> The output file is truncated if S3 serves a stale copy of the temporary INode.  In copyOutput, step 1 above, it is possible for S3 to return a version of the temporary INode that contains just [block_99, block_81].  In this case, we write this new data to the final output location, and 'lose' block_-101 in the process.  Since we then delete the temporary INode, we've lost all references to the final block of this file and it's orphaned in the S3 bucket.
> This type of consistency error is infrequent but not impossible. We've observed these failures about once a week for one of our large jobs which runs daily and has 200 reduce outputs; so we're seeing an error rate of something like 0.07% per reduce.
> These kind of errors are generally difficult to handle in a system like S3.  We have a few ideas about how to fix this:
> 1) HACK! Sleep during S3OutputStream#close or #flush to wait for S3 to catch up and make these less likely.
> 2) Poll for updated MD5 or INode data in Jets3tFileSystemStore#storeINode until S3 says the INode contents are the same as our local copy.  This could be a config option - "fs.s3.verifyInodeWrites" or something like that.
> 3) Cache INode contents in-process, so we don't have to go back to S3 to ask for the current version of an INode.
> 4) Only write INodes once, when the output stream is closed.  This would basically make S3OutputStream#flush() a no-op.
> 5) Modify the S3FS to somehow version INodes (unclear how we would do this, need some design work).
> 6) Avoid using the S3FS for temporary task attempt files.
> 7) Avoid using the S3FS completely.
> We wanted to get some guidance from the community before we went down any of these paths.  Has anyone seen this issue?  Any other suggested workarounds?  We at tracksimple are willing to invest some time in fixing this and (of course) contributing our fix back, but we wanted to get an 'ack' from others before we try anything crazy :-).
> I've attached a test app if anyone wants to try and reproduce this themselves.  It takes a while to run (depending on the 'weather' in S3 right now), but should eventually detect a consistency 'error' that manifests itself as a truncated file.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.