[jira] [Created] (HADOOP-16526) Support LDAP authenticaition (bind) via GSSAPI
Todd Lipcon created HADOOP-16526: Summary: Support LDAP authenticaition (bind) via GSSAPI Key: HADOOP-16526 URL: https://issues.apache.org/jira/browse/HADOOP-16526 Project: Hadoop Common Issue Type: Improvement Components: security Reporter: Todd Lipcon Currently the LDAP group mapping provider only supports simple (user/password) authentication. In some cases it's more convenient to use GSSAPI (kerberos) authentication here, particularly when the server doing the mapping is already using a keytab provided by the same instance (eg IPA or AD). We should provide a configuration to turn on GSSAPI and put the right UGI 'doAs' calls in place to ensure an appropriate Subject in those calls. -- This message was sent by Atlassian Jira (v8.3.2#803003) - To unsubscribe, e-mail: common-dev-unsubscr...@hadoop.apache.org For additional commands, e-mail: common-dev-h...@hadoop.apache.org
[jira] [Created] (HADOOP-16525) LDAP group mapping should include primary posix group
Todd Lipcon created HADOOP-16525: Summary: LDAP group mapping should include primary posix group Key: HADOOP-16525 URL: https://issues.apache.org/jira/browse/HADOOP-16525 Project: Hadoop Common Issue Type: Improvement Reporter: Todd Lipcon Assignee: Todd Lipcon When configuring LdapGroupsMapping against FreeIPA, the current implementation searches for groups which have the user listed as a member. This catches all "secondary" groups but misses the user's primary group (typically the same name as their username). We should include a search for a group matching the user's primary gidNumber in the group search. -- This message was sent by Atlassian Jira (v8.3.2#803003) - To unsubscribe, e-mail: common-dev-unsubscr...@hadoop.apache.org For additional commands, e-mail: common-dev-h...@hadoop.apache.org
[jira] [Created] (HADOOP-16179) hadoop-common pom should not depend on kerb-simplekdc
Todd Lipcon created HADOOP-16179: Summary: hadoop-common pom should not depend on kerb-simplekdc Key: HADOOP-16179 URL: https://issues.apache.org/jira/browse/HADOOP-16179 Project: Hadoop Common Issue Type: Improvement Reporter: Todd Lipcon The hadoop-common pom currently has a dependency on kerb-simplekdc. In fact, the only classes used from Kerby are in kerb-core and kerb-util (which is a transitive dependency frmo kerb-core). Depending on kerb-simplekdc pulls a bunch of other unnecessary classes into the hadoop-common classpath. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: common-dev-unsubscr...@hadoop.apache.org For additional commands, e-mail: common-dev-h...@hadoop.apache.org
[jira] [Created] (HADOOP-16011) OsSecureRandom very slow compared to other SecureRandom implementations
Todd Lipcon created HADOOP-16011: Summary: OsSecureRandom very slow compared to other SecureRandom implementations Key: HADOOP-16011 URL: https://issues.apache.org/jira/browse/HADOOP-16011 Project: Hadoop Common Issue Type: Bug Components: security Reporter: Todd Lipcon In looking at performance of a workload which creates a lot of short-lived remote connections to a secured DN, [~philip] and I found very high system CPU usage. We tracked it down to reads from /dev/random, which are incurred by the DN using CryptoCodec.generateSecureRandom to generate a transient session key and IV for AES encryption. In the case that the OpenSSL codec is not enabled, the above code falls through to the JDK SecureRandom implementation, which performs reasonably. However, OpenSSLCodec defaults to using OsSecureRandom, which reads all random data from /dev/random rather than doing something more efficient like initializing a CSPRNG from a small seed. I wrote a simple JMH benchmark to compare various approaches when running with concurrency 10: testHadoop - using CryptoCodec testNewSecureRandom - using 'new SecureRandom()' each iteration testSha1PrngNew - using the SHA1PRNG explicitly, new instance each iteration testSha1PrngShared - using a single shared instance of SHA1PRNG testSha1PrngThread - using a thread-specific instance of SHA1PRNG {code:java} Benchmark Mode CntScore Error Units MyBenchmark.testHadoop thrpt 1293.000 ops/s [with libhadoop.so] MyBenchmark.testHadoop thrpt461515.697 ops/s [without libhadoop.so] MyBenchmark.testNewSecureRandom thrpt 43413.640 ops/s MyBenchmark.testSha1PrngNew thrpt395515.000 ops/s MyBenchmark.testSha1PrngShared thrpt164488.713 ops/s MyBenchmark.testSha1PrngThread thrpt 4295123.210 ops/s {code} In other words, the presence of the OpenSSL acceleration slows down this code path by 356x. And, compared to the optimal (thread-local Sha1Prng) it's 3321x slower. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: common-dev-unsubscr...@hadoop.apache.org For additional commands, e-mail: common-dev-h...@hadoop.apache.org
Re: [DISCUSS] Hadoop RPC encryption performance improvements
One possibility (which we use in Kudu) is to use SSL for encryption but with a self-signed certificate, maintaining the existing SASL/GSSAPI handshake for authentication. The one important bit here, security wise, is to implement channel binding (RFC 5056 and RFC 5929) to prevent against MITMs. The description of the Kudu protocol is here: https://github.com/apache/kudu/blob/master/docs/design-docs/rpc.md#wire-protocol If implemented correctly, this provides TLS encryption (with all of its performance and security benefits) without requiring the user to deploy a custom cert. -Todd On Thu, Nov 1, 2018 at 7:14 PM Konstantin Shvachko wrote: > Hi Wei-Chiu, > > Thanks for starting the thread and summarizing the problem. Sorry for slow > response. > We've been looking at the encrypted performance as well and are interested > in this effort. > We ran some benchmarks locally. Our benchmarks also showed substantial > penalty for turning on wire encryption on rpc. > Although it was less drastic - more in the range of -40%. But we ran a > different benchmark NNThroughputBenchmark, and we ran it on 2.6 last year. > Could have published the results, but need to rerun on more recent > versions. > > Three points from me on this discussion: > > 1. We should settle on the benchmarking tools. > For development RPCCallBenchmark is good as it measures directly the > improvement on the RPC layer. But for external consumption it is more > important to know about e.g. NameNode RPCs performance. So we probably > should run both benchmarks. > 2. SASL vs SSL. > Since current implementation is based on SASL, I think it would make sense > to make improvements in this direction. I assume switching to SSL would > require changes in configuration. Not sure if it will be compatible, since > we don't have the details. At this point I would go with HADOOP-10768. > Given all (Daryn's) concerns are addressed. > 3. Performance improvement expectations. > Ideally we want to have < 10% penalty for encrypted communication. Anything > over 30% will probably have very limited usability. And there is the gray > area in between, which could be mitigated by allowing mixed encrypted and > un-encrypted RPCs on the single NameNode like in HDFS-13566. > > Thanks, > --Konstantin > > On Wed, Oct 31, 2018 at 7:39 AM Daryn Sharp > wrote: > > > Various KMS tasks have been delaying my RPC encryption work – which is > 2nd > > on TODO list. It's becoming a top priority for us so I'll try my best to > > get a preliminary netty server patch (sans TLS) up this week if that > helps. > > > > The two cited jiras had some critical flaws. Skimming my comments, both > > use blocking IO (obvious nonstarter). HADOOP-10768 is a hand rolled > > TLS-like encryption which I don't feel is something the community can or > > should maintain from a security standpoint. > > > > Daryn > > > > On Wed, Oct 31, 2018 at 8:43 AM Wei-Chiu Chuang > > wrote: > > > > > Ping. Any one? Cloudera is interested in moving forward with the RPC > > > encryption improvements, but I just like to get a consensus which > > approach > > > to go with. > > > > > > Otherwise I'll pick HADOOP-10768 since it's ready for commit, and I've > > > spent time on testing it. > > > > > > On Thu, Oct 25, 2018 at 11:04 AM Wei-Chiu Chuang > > > wrote: > > > > > > > Folks, > > > > > > > > I would like to invite all to discuss the various Hadoop RPC > encryption > > > > performance improvements. As you probably know, Hadoop RPC encryption > > > > currently relies on Java SASL, and have _really_ bad performance (in > > > terms > > > > of number of RPCs per second, around 15~20% of the one without SASL) > > > > > > > > There have been some attempts to address this, most notably, > > HADOOP-10768 > > > > <https://issues.apache.org/jira/browse/HADOOP-10768> (Optimize > Hadoop > > > RPC > > > > encryption performance) and HADOOP-13836 > > > > <https://issues.apache.org/jira/browse/HADOOP-13836> (Securing > Hadoop > > > RPC > > > > using SSL). But it looks like both attempts have not been > progressing. > > > > > > > > During the recent Hadoop contributor meetup, Daryn Sharp mentioned > he's > > > > working on another approach that leverages Netty for its SSL > > encryption, > > > > and then integrate Netty with Hadoop RPC so that Hadoop RPC > > automatically > > > > benefits from netty's SSL encryption performance. > > > > > > > > So there are at least 3 attempts to address this issue as I see it. > Do > > we > > > > have a consensus that: > > > > 1. this is an important problem > > > > 2. which approach we want to move forward with > > > > > > > > -- > > > > A very happy Hadoop contributor > > > > > > > > > > > > > -- > > > A very happy Hadoop contributor > > > > > > > > > -- > > > > Daryn > > > -- Todd Lipcon Software Engineer, Cloudera
Re: [DISCUSS]: securing ASF Hadoop releases out of the box
gs to think about > > > > * docs explaining IN CAPITAL LETTERS why you need to lock down your > > cluster to a private subnet or use Kerberos > > * Anything which can be done to make Kerberos easier (?). I see > there are > > some oustanding patches for HADOOP-12649 which need review, but what > else? > > > > Could we have Hadoop determine when it's coming up on an open > network and > > start warning? And how? > > > > At the very least, single node hadoop should be locked down. You > shouldn't > > have to bring up kerberos to run it like that. And for more > sophisticated > > multinode deployments, should the scripts refuse to work without > kerberos > > unless you pass in some argument like "--Dinsecure-clusters- > permitted" > > > > Any other ideas? > > > > > > > - > > To unsubscribe, e-mail: common-dev-unsubscr...@hadoop.apache.org > > For additional commands, e-mail: common-dev-h...@hadoop.apache.org > > > > > > > -- Todd Lipcon Software Engineer, Cloudera
[jira] [Created] (HADOOP-15566) Remove HTrace support
Todd Lipcon created HADOOP-15566: Summary: Remove HTrace support Key: HADOOP-15566 URL: https://issues.apache.org/jira/browse/HADOOP-15566 Project: Hadoop Common Issue Type: Improvement Reporter: Todd Lipcon The HTrace incubator project has voted to retire itself and won't be making further releases. The Hadoop project currently has various hooks with HTrace. It seems in some cases (eg HDFS-13702) these hooks have had measurable performance overhead. Given these two factors, I think we should consider removing the HTrace integration. If there is someone willing to do the work, replacing it with OpenTracing might be a better choice since there is an active community. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: common-dev-unsubscr...@hadoop.apache.org For additional commands, e-mail: common-dev-h...@hadoop.apache.org
[jira] [Created] (HADOOP-15564) Classloading Shell should not run a subprocess
Todd Lipcon created HADOOP-15564: Summary: Classloading Shell should not run a subprocess Key: HADOOP-15564 URL: https://issues.apache.org/jira/browse/HADOOP-15564 Project: Hadoop Common Issue Type: Improvement Components: util Affects Versions: 3.0.0 Reporter: Todd Lipcon The 'Shell' class has a static member isSetsidSupported which, in order to initialize, forks out a subprocess. Various other parts of the code reference Shell.WINDOWS. For example, the StringUtils class has such a reference. This means that, during startup, a seemingly fast call like Configuration.getBoolean() ends up class-loading StringUtils, which class-loads Shell, which forks out a subprocess. I couldn't measure any big improvement by fixing this, but seemed surprising to say the least. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: common-dev-unsubscr...@hadoop.apache.org For additional commands, e-mail: common-dev-h...@hadoop.apache.org
[jira] [Created] (HADOOP-15557) Crypto streams should not crash when mis-used
Todd Lipcon created HADOOP-15557: Summary: Crypto streams should not crash when mis-used Key: HADOOP-15557 URL: https://issues.apache.org/jira/browse/HADOOP-15557 Project: Hadoop Common Issue Type: Improvement Components: fs Affects Versions: 3.2.0 Reporter: Todd Lipcon In general, the non-positional read APIs for streams in Hadoop Common are meant to be used by only a single thread at a time. It would not make much sense to have concurrent multi-threaded access to seek+read because they modify the stream's file position. Multi-threaded access on input streams can be done using positional read APIs. Multi-threaded access on output streams probably never makes sense. In the case of DFSInputStream, the positional read APIs are marked synchronized, so that even when misused, no strange exceptions are thrown. The results are just somewhat undefined in that it's hard for a thread to know which position was read from. However, when running on an encrypted file system, the results are much worse: since CryptoInputStream's read methods are not marked synchronized, the caller can get strange ByteBuffer exceptions or even a JVM crash due to concurrent use and free of underlying OpenSSL Cipher buffers. The crypto stream wrappers should be made more resilient to such misuse, for example by: (a) making the read methods safer by making them synchronized (so they have the same behavior as DFSInputStream) or (b) trying to detect concurrent access to these methods and throwing ConcurrentModificationException so that the user is alerted to their probable misuse. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: common-dev-unsubscr...@hadoop.apache.org For additional commands, e-mail: common-dev-h...@hadoop.apache.org
[jira] [Created] (HADOOP-15554) Improve JIT performance for Configuration parsing
Todd Lipcon created HADOOP-15554: Summary: Improve JIT performance for Configuration parsing Key: HADOOP-15554 URL: https://issues.apache.org/jira/browse/HADOOP-15554 Project: Hadoop Common Issue Type: Improvement Components: conf, performance Affects Versions: 3.0.0 Reporter: Todd Lipcon Assignee: Todd Lipcon In investigating a performance regression for small tasks between Hadoop 2 and Hadoop 3, we found that the amount of time spent in JIT was significantly higher. Using jitwatch we were able to determine that, due to a combination of switching from DOM to SAX style parsing and just having more configuration key/value pairs, Configuration.loadResource is now getting compiled with the C2 compiler and taking quite some time. Breaking that very large function up into several smaller ones and eliminating some redundant bits of code improves the JIT performance measurably. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: common-dev-unsubscr...@hadoop.apache.org For additional commands, e-mail: common-dev-h...@hadoop.apache.org
[jira] [Created] (HADOOP-15551) Avoid use of Java8 streams in Configuration.addTags
Todd Lipcon created HADOOP-15551: Summary: Avoid use of Java8 streams in Configuration.addTags Key: HADOOP-15551 URL: https://issues.apache.org/jira/browse/HADOOP-15551 Project: Hadoop Common Issue Type: Improvement Components: performance Affects Versions: 3.2 Reporter: Todd Lipcon Assignee: Todd Lipcon Configuration.addTags oddly uses Arrays.stream instead of a more conventional mechanism. When profiling a simple program that uses Configuration, I found that addTags was taking tens of millis of CPU to do very little work the first time it's called, accounting for ~8% of total profiler samples in my program. {code} [9] 4.52% 253 self: 0.00% 0 java/lang/invoke/MethodHandleNatives.linkCallSite [9] 3.71% 208 self: 0.00% 0 java/lang/invoke/MethodHandleNatives.linkMethodHandleConstant {code} I don't know much about the implementation details of the Streams stuff, but it seems it's probably meant more for cases with very large arrays or somesuch. Switching to a normal Set.addAll() call eliminates this from the profile. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: common-dev-unsubscr...@hadoop.apache.org For additional commands, e-mail: common-dev-h...@hadoop.apache.org
[jira] [Created] (HADOOP-15550) Avoid static initialization of ObjectMappers
Todd Lipcon created HADOOP-15550: Summary: Avoid static initialization of ObjectMappers Key: HADOOP-15550 URL: https://issues.apache.org/jira/browse/HADOOP-15550 Project: Hadoop Common Issue Type: Bug Components: performance Affects Versions: 3.2.0 Reporter: Todd Lipcon Assignee: Todd Lipcon Various classes statically initialize an ObjectMapper READER instance. This ends up doing a bunch of class-loading of Jackson libraries that can add up to a fair amount of CPU, even if the reader ends up not being used. This is particularly the case with WebHdfsFileSystem, which is class-loaded by a serviceloader even when unused in a particular job. We should lazy-init these members instead of doing so as a static class member. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: common-dev-unsubscr...@hadoop.apache.org For additional commands, e-mail: common-dev-h...@hadoop.apache.org
[jira] [Created] (HADOOP-15549) Upgrade to commons-configuration 2.1 regresses task CPU consumption
Todd Lipcon created HADOOP-15549: Summary: Upgrade to commons-configuration 2.1 regresses task CPU consumption Key: HADOOP-15549 URL: https://issues.apache.org/jira/browse/HADOOP-15549 Project: Hadoop Common Issue Type: Bug Components: metrics Affects Versions: 3.0.2 Reporter: Todd Lipcon Assignee: Todd Lipcon HADOOP-13660 upgraded from commons-configuration 1.x to 2.x. commons-configuration is used when parsing the metrics configuration properties file. The new builder API used in the new version apparently makes use of a bunch of very bloated reflection and classloading nonsense to achieve the same goal, and this results in a regression of >100ms of CPU time as measured by a program which simply initializes DefaultMetricsSystem. This isn't a big deal for long-running daemons, but for MR tasks which might only run a few seconds on poorly-tuned jobs, this can be noticeable. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: common-dev-unsubscr...@hadoop.apache.org For additional commands, e-mail: common-dev-h...@hadoop.apache.org
[jira] [Resolved] (HADOOP-9545) Improve logging in ActiveStandbyElector
[ https://issues.apache.org/jira/browse/HADOOP-9545?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Todd Lipcon resolved HADOOP-9545. - Resolution: Won't Fix > Improve logging in ActiveStandbyElector > --- > > Key: HADOOP-9545 > URL: https://issues.apache.org/jira/browse/HADOOP-9545 > Project: Hadoop Common > Issue Type: Improvement > Components: auto-failover, ha >Affects Versions: 2.1.0-beta > Reporter: Todd Lipcon >Assignee: Todd Lipcon >Priority: Minor > > The ActiveStandbyElector currently logs a lot of stuff at DEBUG level which > would be useful for troubleshooting. We've seen one instance in the wild of a > ZKFC thinking it should be in standby state when in fact it won the election, > but the logging is insufficient to understand why. I'd like to bump most of > the existing DEBUG logs to INFO and add some additional logs as well. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: common-dev-unsubscr...@hadoop.apache.org For additional commands, e-mail: common-dev-h...@hadoop.apache.org
[jira] [Resolved] (HADOOP-10859) Native implementation of java Checksum interface
[ https://issues.apache.org/jira/browse/HADOOP-10859?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Todd Lipcon resolved HADOOP-10859. -- Resolution: Won't Fix No plans to work on this. > Native implementation of java Checksum interface > > > Key: HADOOP-10859 > URL: https://issues.apache.org/jira/browse/HADOOP-10859 > Project: Hadoop Common > Issue Type: Improvement > Reporter: Todd Lipcon > Assignee: Todd Lipcon >Priority: Minor > > Some parts of our code such as IFileInputStream/IFileOutputStream use the > java Checksum interface to calculate/verify checksums. Currently we don't > have a native implementation of these. For CRC32C in particular, we can get a > very big speedup with a native implementation. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: common-dev-unsubscr...@hadoop.apache.org For additional commands, e-mail: common-dev-h...@hadoop.apache.org
Version control link on webpage
https://hadoop.apache.org/version_control.html Still seems to point to SVN instead of git. Is the site itself still versioned from the SVN repository or has that also transferred to git so we can fix this? -Todd -- Todd Lipcon Software Engineer, Cloudera
Re: Git repo ready to use
, 2014 at 4:21 PM, Karthik Kambatla ka...@cloudera.com wrote: Thanks Giri and Ted for fixing the builds. On Thu, Aug 28, 2014 at 9:49 AM, Ted Yu yuzhih...@gmail.com wrote: Charles: QA build is running for your JIRA: https://builds.apache.org/job/PreCommit-hdfs-Build/7828/parameters/ Cheers On Thu, Aug 28, 2014 at 9:41 AM, Charles Lamb cl...@cloudera.com wrote: On 8/28/2014 12:07 PM, Giridharan Kesavan wrote: Fixed all the 3 pre-commit buids. test-patch's git reset --hard is removing the patchprocess dir, so moved it off the workspace. Thanks Giri. Should I resubmit HDFS-6954's patch? I've gotten 3 or 4 jenkins messages that indicated the problem so something is resubmitting, but now that you've fixed it, should I resubmit it again? Charles -- Alejandro -- Todd Lipcon Software Engineer, Cloudera
Re: migrating private branches to the new git repo
On Tue, Sep 2, 2014 at 2:38 PM, Andrew Wang andrew.w...@cloudera.com wrote: Not to derail the conversation, but if CHANGES.txt is making backports more annoying, why don't we get rid of it? It seems like we should be able to generate it via a JIRA query, and git log can also be used for a quick check (way faster than svn log). +1, I've always found CHANGES.txt to be a big pain in the butt, and often it gets incorrect, too. On Tue, Sep 2, 2014 at 12:38 PM, Steve Loughran ste...@hortonworks.com wrote: I've now done my first commits; one into trunk (10373), one into branch-2 and cherry picked (fix in hadoop-common-project/hadoop-common/src/main/native/README ; no JIRA). I made an initial attempt to cherry pick the HADOOP-10373 patch from trunk into branch-2, with CHANGES.TXT being a dramatic enough change that it takes human intervention to patch. implication 1. committing to branch-2 with changes.txt in the same commit followed by a cherry pick forwards works. 2. committing to trunk only backports reliably if the changes.txt files are patched in a separate commit This is no different from SVN, except that an svn merge used different commands. I have not tried the git format-patch/git am option, which would be: 1. -use git am -3 to apply the patch to the HEAD of both branch-2 and trunk 2. -patch changes.txt in each branch, then either commit separately 3. -or try and amend latest commit for the patches #3 seems appealing, but it'd make the diff on the two branches different. On 2 September 2014 19:01, Andrew Wang andrew.w...@cloudera.com wrote: This is basically what I did, make patches of each of my branches and then reapply to the new trunk. One small recommendation would be to make the remote named apache rather than asflive so it's consistent with the GitAndHadoop wikipage. IMO naming branches with a / (e.g. live/trunk) is also kind of ambiguous, since it's the same syntax used to specify a remote. It seems there can also be difficulties with directory and filenames. Somewhat related, it'd be nice to update the GitAndHadoop instructions on how to generate a patch using git-format-patch. I've been using plain old git diff for a while, but format-patch seems better. It'd be especially nice if a recommended .gitconfig section was made available :) I plan to play with format-patch some in the near future and might do this myself, but if any git gurus already have this ready to go, feel free to edit. On Tue, Sep 2, 2014 at 4:10 AM, Steve Loughran ste...@hortonworks.com wrote: Now that hadoop is using git, I'm migrating my various work-in-progress branches to the new commit tree 1. This is the process I've written up for using git format-patch then git am to export the patch sequence and merge it in, then rebasing onto trunk to finally get in sync https://wiki.apache.org/hadoop/MigratingPrivateGitBranches 2. The Git and hadoop docs cover git graft: https://wiki.apache.org/hadoop/GitAndHadoop#Grafts_for_complete_project_history I'm not sure if/how that relates Is there any easier way than what I've described for doing the move? -- CONFIDENTIALITY NOTICE NOTICE: This message is intended for the use of the individual or entity to which it is addressed and may contain information that is confidential, privileged and exempt from disclosure under applicable law. If the reader of this message is not the intended recipient, you are hereby notified that any printing, copying, dissemination, distribution, disclosure or forwarding of this communication is strictly prohibited. If you have received this communication in error, please contact the sender immediately and delete it from your system. Thank You. -- CONFIDENTIALITY NOTICE NOTICE: This message is intended for the use of the individual or entity to which it is addressed and may contain information that is confidential, privileged and exempt from disclosure under applicable law. If the reader of this message is not the intended recipient, you are hereby notified that any printing, copying, dissemination, distribution, disclosure or forwarding of this communication is strictly prohibited. If you have received this communication in error, please contact the sender immediately and delete it from your system. Thank You. -- Todd Lipcon Software Engineer, Cloudera
Re: Updates on migration to git
of this communication is strictly prohibited. If you have received this communication in error, please contact the sender immediately and delete it from your system. Thank You. -- Mobile -- CONFIDENTIALITY NOTICE NOTICE: This message is intended for the use of the individual or entity to which it is addressed and may contain information that is confidential, privileged and exempt from disclosure under applicable law. If the reader of this message is not the intended recipient, you are hereby notified that any printing, copying, dissemination, distribution, disclosure or forwarding of this communication is strictly prohibited. If you have received this communication in error, please contact the sender immediately and delete it from your system. Thank You. -- Todd Lipcon Software Engineer, Cloudera
[jira] [Created] (HADOOP-10882) Move DirectBufferPool into common util
Todd Lipcon created HADOOP-10882: Summary: Move DirectBufferPool into common util Key: HADOOP-10882 URL: https://issues.apache.org/jira/browse/HADOOP-10882 Project: Hadoop Common Issue Type: Task Components: util Affects Versions: 2.6.0 Reporter: Todd Lipcon Assignee: Todd Lipcon Priority: Minor MAPREDUCE-2841 uses a direct buffer pool to pass data back and forth between native and Java code. The branch has an implementation which appears to be derived from the one in HDFS. Instead of copy-pasting, we should move the HDFS DirectBufferPool into Common so that MR can make use of it. -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Created] (HADOOP-10859) Native implementation of java Checksum interface
Todd Lipcon created HADOOP-10859: Summary: Native implementation of java Checksum interface Key: HADOOP-10859 URL: https://issues.apache.org/jira/browse/HADOOP-10859 Project: Hadoop Common Issue Type: Improvement Reporter: Todd Lipcon Assignee: Todd Lipcon Priority: Minor Some parts of our code such as IFileInputStream/IFileOutputStream use the java Checksum interface to calculate/verify checksums. Currently we don't have a native implementation of these. For CRC32C in particular, we can get a very big speedup with a native implementation. -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Created] (HADOOP-10855) Allow Text to be read with a known length
Todd Lipcon created HADOOP-10855: Summary: Allow Text to be read with a known length Key: HADOOP-10855 URL: https://issues.apache.org/jira/browse/HADOOP-10855 Project: Hadoop Common Issue Type: Improvement Components: io Affects Versions: 2.6.0 Reporter: Todd Lipcon Assignee: Todd Lipcon Priority: Minor For the native task work (MAPREDUCE-2841) it is useful to be able to store strings in a different fashion than the default (varint-prefixed) serialization. We should provide a read method in Text which takes an already-known length to support this use case while still providing Text objects back to the user. -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Created] (HADOOP-10288) Explicit reference to Log4JLogger breaks non-log4j users
Todd Lipcon created HADOOP-10288: Summary: Explicit reference to Log4JLogger breaks non-log4j users Key: HADOOP-10288 URL: https://issues.apache.org/jira/browse/HADOOP-10288 Project: Hadoop Common Issue Type: Bug Components: util Affects Versions: 2.4.0 Reporter: Todd Lipcon Assignee: Todd Lipcon In HttpRequestLog, we make an explicit reference to the Log4JLogger class for an instanceof check. If the log4j implementation isn't actually on the classpath, the instanceof check throws NoClassDefFoundError instead of returning false. This means that dependent projects that don't use log4j can no longer embed HttpServer -- typically this is an issue when they use MiniDFSCluster as part of their testing. -- This message was sent by Atlassian JIRA (v6.1.5#6160)
[jira] [Resolved] (HADOOP-10199) Precommit Admin build is not running because no previous successful build is available
[ https://issues.apache.org/jira/browse/HADOOP-10199?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Todd Lipcon resolved HADOOP-10199. -- Resolution: Fixed Hadoop Flags: Reviewed Precommit Admin build is not running because no previous successful build is available -- Key: HADOOP-10199 URL: https://issues.apache.org/jira/browse/HADOOP-10199 Project: Hadoop Common Issue Type: Bug Reporter: Brock Noland Assignee: Brock Noland Priority: Blocker Attachments: HADOOP-10199.patch It seems at some point the builds started failing for an unknown reason and eventually the last successful was rolled off. At that point the precommit builds started failing because they pull an artifact from the last successful build. -- This message was sent by Atlassian JIRA (v6.1.5#6160)
[jira] [Resolved] (HADOOP-10200) Fix precommit script patch_tested.txt fallback option
[ https://issues.apache.org/jira/browse/HADOOP-10200?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Todd Lipcon resolved HADOOP-10200. -- Resolution: Fixed Hadoop Flags: Reviewed Fix precommit script patch_tested.txt fallback option - Key: HADOOP-10200 URL: https://issues.apache.org/jira/browse/HADOOP-10200 Project: Hadoop Common Issue Type: Bug Reporter: Brock Noland Assignee: Brock Noland Attachments: HADOOP-10200.patch HADOOP-10199 created a fallback option when there is successful artifact. However that fallback option used the jenkins lastBuild build indicator. It appears that does not mean the last completed build, but strictly the last build, which in this context is the current build. The current build is running so it doesn't have any artifacts. -- This message was sent by Atlassian JIRA (v6.1.5#6160)
[jira] [Resolved] (HADOOP-9765) Precommit Admin job chokes on issues without an attachment
[ https://issues.apache.org/jira/browse/HADOOP-9765?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Todd Lipcon resolved HADOOP-9765. - Resolution: Fixed Hadoop Flags: Reviewed Precommit Admin job chokes on issues without an attachment -- Key: HADOOP-9765 URL: https://issues.apache.org/jira/browse/HADOOP-9765 Project: Hadoop Common Issue Type: Bug Components: build Reporter: Brock Noland Assignee: Brock Noland Attachments: HADOOP-9765.patch, HADOOP-9765.patch Check out this file: https://builds.apache.org/job/PreCommit-Admin/lastSuccessfulBuild/artifact/patch_tested.txt It has corrupt data: {noformat} HIVE-4877HDFS-5010,12593214 HIVE-4877HBASE-8693,12593082 HIVE-4877YARN-919,12593107 YARN-905,12593225 HIVE-4877HBASE-8752,12588069 {noformat} which resulted in the Hive precommit job being called with the ISSUE_NUM of 5010, 8693, 919, and 8752. Looking at the script and some output, I pulled from the last run, it looks like it gets hosed up when there is a JIRA which is PA but doesn't have an attachment (as ZK-1402 is currently sitting). For example: This is the bad data the script is encountering: {noformat} $ grep -A 2 'ZOOKEEPER-1402' patch_available2.elements ZOOKEEPER-1402 HBASE-8348 id=12592318 {noformat} This is where it screws up: {noformat} $ awk '{ printf %s, $0 }' patch_available2.elements | sed -e s/\W*id=\/,/g | perl -pe s/\/\n/g | grep ZOOKEEPER-1402 ZOOKEEPER-1402HBASE-8348 ,12592318 {noformat} -- This message was sent by Atlassian JIRA (v6.1#6144)
Re: Next releases
On Mon, Nov 11, 2013 at 2:57 PM, Colin McCabe cmcc...@alumni.cmu.eduwrote: To be honest, I'm not aware of anything in 2.2.1 that shouldn't be there. However, I have only been following the HDFS and common side of things so I may not have the full picture. Arun, can you give a specific example of something you'd like to blow away? I agree with Colin. If we've been backporting things into a patch release (third version component) which don't belong, we should explicitly call out those patches, so we can learn from our mistakes and have a discussion about what belongs. Otherwise we'll just end up doing it again. Saying there were a few mistakes, so let's reset back a bunch of backport work seems like a baby-with-the-bathwater situation. Todd
[jira] [Created] (HADOOP-9908) Fix NPE when versioninfo properties file is missing
Todd Lipcon created HADOOP-9908: --- Summary: Fix NPE when versioninfo properties file is missing Key: HADOOP-9908 URL: https://issues.apache.org/jira/browse/HADOOP-9908 Project: Hadoop Common Issue Type: Bug Components: util Affects Versions: 2.1.0-beta, 3.0.0 Reporter: Todd Lipcon Assignee: Todd Lipcon Attachments: hadoop-9908.txt When running tests in Eclipse I ran into an NPE in VersionInfo since the version info properties file didn't properly make it to the classpath. This is because getResourceAsStream can return null if the file is not found. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Resolved] (HADOOP-8336) LocalFileSystem Does not seek to the correct location when Checksumming is off.
[ https://issues.apache.org/jira/browse/HADOOP-8336?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Todd Lipcon resolved HADOOP-8336. - Resolution: Duplicate Resolved as duplicate of HADOOP-9307, since I'm pretty sure that solved this issue. LocalFileSystem Does not seek to the correct location when Checksumming is off. --- Key: HADOOP-8336 URL: https://issues.apache.org/jira/browse/HADOOP-8336 Project: Hadoop Common Issue Type: Bug Components: fs Reporter: Elliott Clark Assignee: Todd Lipcon Attachments: branch-1-test.txt Hbase was seeing an issue when trying to read data from a local filesystem instance with setVerifyChecksum(false). On debugging into it, the seek on the file was seeking to the checksum block index, but since checksumming was off that was the incorrect location. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-9898) Set SO_KEEPALIVE on all our sockets
Todd Lipcon created HADOOP-9898: --- Summary: Set SO_KEEPALIVE on all our sockets Key: HADOOP-9898 URL: https://issues.apache.org/jira/browse/HADOOP-9898 Project: Hadoop Common Issue Type: Bug Components: ipc, net Affects Versions: 3.0.0 Reporter: Todd Lipcon Priority: Minor We recently saw an issue where network issues between slaves and the NN caused ESTABLISHED TCP connections to pile up and leak on the NN side. It looks like the RST packets were getting dropped, which meant that the client thought the connections were closed, while they hung open forever on the server. Setting the SO_KEEPALIVE option on our sockets would prevent this kind of leak from going unchecked. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-9707) Fix register lists for crc32c inline assembly
Todd Lipcon created HADOOP-9707: --- Summary: Fix register lists for crc32c inline assembly Key: HADOOP-9707 URL: https://issues.apache.org/jira/browse/HADOOP-9707 Project: Hadoop Common Issue Type: Bug Components: util Affects Versions: 3.0.0, 2.1.0-beta Reporter: Todd Lipcon Assignee: Todd Lipcon Priority: Minor The inline assembly used for the crc32 instructions has an incorrect clobber list: the computed CRC values are in-out variables and thus need to use the matching constraint syntax in the clobber list. This doesn't seem to cause a problem now in Hadoop, but may break in a different compiler version which allocates registers differently, or may break when the same code is used in another context. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-9618) Add thread which detects JVM pauses
Todd Lipcon created HADOOP-9618: --- Summary: Add thread which detects JVM pauses Key: HADOOP-9618 URL: https://issues.apache.org/jira/browse/HADOOP-9618 Project: Hadoop Common Issue Type: New Feature Components: util Affects Versions: 3.0.0 Reporter: Todd Lipcon Assignee: Todd Lipcon Often times users struggle to understand what happened when a long JVM pause (GC or otherwise) causes things to malfunction inside a Hadoop daemon. For example, a long GC pause while logging an edit to the QJM may cause the edit to timeout, or a long GC pause may make other IPCs to the NameNode timeout. We should add a simple thread which loops on 1-second sleeps, and if the sleep ever takes significantly longer than 1 second, log a WARN. This will make GC pauses obvious in logs. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-9608) ZKFC should abort if it sees an unrecognized NN become active
Todd Lipcon created HADOOP-9608: --- Summary: ZKFC should abort if it sees an unrecognized NN become active Key: HADOOP-9608 URL: https://issues.apache.org/jira/browse/HADOOP-9608 Project: Hadoop Common Issue Type: Bug Components: ha Affects Versions: 3.0.0 Reporter: Todd Lipcon We recently had an issue where one NameNode and ZKFC was updated to a new configuration/IP address but the ZKFC on the other node was not rebooted. Then, next time a failover occurred, the second ZKFC was not able to become active because the data in the ActiveBreadCrumb didn't match the data in its own configuration: {code} org.apache.hadoop.ha.ActiveStandbyElector: Exception handling the winning of election java.lang.IllegalArgumentException: Unable to determine service address for namenode '' {code} To prevent this from happening, whenever the ZKFC sees a new NN become active, it should check that it's properly able to instantiate a ServiceTarget for it, and if not, abort (since this ZKFC wouldn't be able to handle a failover successfully) -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-9601) Support native CRC on byte arrays
Todd Lipcon created HADOOP-9601: --- Summary: Support native CRC on byte arrays Key: HADOOP-9601 URL: https://issues.apache.org/jira/browse/HADOOP-9601 Project: Hadoop Common Issue Type: Improvement Components: performance, util Affects Versions: 3.0.0 Reporter: Todd Lipcon When we first implemented the Native CRC code, we only did so for direct byte buffers, because these correspond directly to native heap memory and thus make it easy to access via JNI. We'd generally assumed that accessing byte[] arrays from JNI was not efficient enough, but now that I know more about JNI I don't think that's true -- we just need to make sure that the critical sections where we lock the buffers are short. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-9545) Improve logging in ActiveStandbyElector
Todd Lipcon created HADOOP-9545: --- Summary: Improve logging in ActiveStandbyElector Key: HADOOP-9545 URL: https://issues.apache.org/jira/browse/HADOOP-9545 Project: Hadoop Common Issue Type: Improvement Components: auto-failover, ha Affects Versions: 2.0.5-beta Reporter: Todd Lipcon Assignee: Todd Lipcon Priority: Minor The ActiveStandbyElector currently logs a lot of stuff at DEBUG level which would be useful for troubleshooting. We've seen one instance in the wild of a ZKFC thinking it should be in standby state when in fact it won the election, but the logging is insufficient to understand why. I'd like to bump most of the existing DEBUG logs to INFO and add some additional logs as well. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-9420) Add percentile or max metric for rpcQueueTime, processing time
Todd Lipcon created HADOOP-9420: --- Summary: Add percentile or max metric for rpcQueueTime, processing time Key: HADOOP-9420 URL: https://issues.apache.org/jira/browse/HADOOP-9420 Project: Hadoop Common Issue Type: Bug Components: ipc, metrics Affects Versions: 2.0.3-alpha Reporter: Todd Lipcon Currently, we only export averages for rpcQueueTime and rpcProcessingTime. These metrics are most useful when looking at timeouts and slow responses, which in my experience are often caused by momentary spikes in load, which won't show up in averages over the 15+ second time intervals often used by metrics systems. We should collect at least the max queuetime and processing time over each interval, or the percentiles if it's not too expensive. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-9399) protoc maven plugin doesn't work on mvn 3.0.2
Todd Lipcon created HADOOP-9399: --- Summary: protoc maven plugin doesn't work on mvn 3.0.2 Key: HADOOP-9399 URL: https://issues.apache.org/jira/browse/HADOOP-9399 Project: Hadoop Common Issue Type: Bug Components: build Affects Versions: 3.0.0 Reporter: Todd Lipcon Assignee: Todd Lipcon Priority: Minor Attachments: hadoop-9399.txt On my machine with mvn 3.0.2, I get a ClassCastException trying to use the maven protoc plugin. The issue seems to be that mvn 3.0.2 sees the ListFile parameter, and doesn't see the generic type argument, and stuffs Strings inside instead. So, we get ClassCastException trying to use the objects as Files. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-9358) Auth failed log should include exception string
Todd Lipcon created HADOOP-9358: --- Summary: Auth failed log should include exception string Key: HADOOP-9358 URL: https://issues.apache.org/jira/browse/HADOOP-9358 Project: Hadoop Common Issue Type: Bug Components: ipc, security Affects Versions: 3.0.0, 2.0.4-beta Reporter: Todd Lipcon Assignee: Todd Lipcon Currently, when authentication fails, we see a WARN message like: {code} 2013-02-28 22:49:03,152 WARN ipc.Server (Server.java:saslReadAndProcess(1056)) - Auth failed for 1.2.3.4:12345:null {code} This is not useful to understand the underlying cause. The WARN entry should additionally include the exception text, eg: {code} 2013-02-28 22:49:03,152 WARN ipc.Server (Server.java:saslReadAndProcess(1056)) - Auth failed for 1.2.3.4:12345:null (GSS initiate failed [Caused by GSSException: Failure unspecified at GSS-API level (Mechanism level: Request is a replay (34))]) {code} -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
Re: [Vote] Merge branch-trunk-win to trunk
mapping, hardlinks, symbolic links, chmod, disk utilization, and process/task management. 3. Added cmd scripts equivalent to existing shell scripts hadoop-daemon.sh, start and stop scripts. 4. Addition of block placement policy implemnation to support cloud enviroment, more specifically Azure. We are very close to wrapping up the work in branch-trunk-win and getting ready for a merge. Currently the merge patch is passing close to 100% of unit tests on Linux. Soon I will call for a vote to merge this branch into trunk. Next steps: 1. Call for vote to merge branch-trunk-win to trunk, when the work completes and precommit build is clean. 2. Start a discussion on adding Jenkins precommit builds on windows and how to integrate that with the existing commit process. Let me know if you have any questions. Regards, Suresh -- http://hortonworks.com/download/ -- http://hortonworks.com/download/ -- Todd Lipcon Software Engineer, Cloudera
[jira] [Created] (HADOOP-9307) BufferedFSInputStream.read returns wrong results after certain seeks
Todd Lipcon created HADOOP-9307: --- Summary: BufferedFSInputStream.read returns wrong results after certain seeks Key: HADOOP-9307 URL: https://issues.apache.org/jira/browse/HADOOP-9307 Project: Hadoop Common Issue Type: Bug Components: fs Affects Versions: 2.0.2-alpha, 1.1.1 Reporter: Todd Lipcon Assignee: Todd Lipcon After certain sequences of seek/read, BufferedFSInputStream can silently return data from the wrong part of the file. Further description in first comment below. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
Re: development environment for hadoop core
Hi Erik, When I started out on Hadoop development, I used to use emacs for most of my development. I eventually saw the light and switched to eclipse with a bunch of emacs keybindings - using an IDE is really handy in Java for functions like find callers of, quick navigation to types, etc. etags gets you part of the way, but I'm pretty sold on eclipse at this point. The other big advantage I found of Eclipse is that the turnaround time on running tests is near-instant - make a change, hit save, and run a unit test in a second or two, instead of waiting 20+sec for maven (even on a non-clean build). That said, for quick fixes or remote debugging work I fall back to vim pretty quickly. -Todd On Tue, Jan 15, 2013 at 3:50 PM, Erik Paulson epaul...@unit1127.com wrote: Hello - I'm curious what Hadoop developers use for their day-to-day hacking on Hadoop. I'm talking changes to the Hadoop libraries and daemons, and not developing Map-Reduce jobs or using using the HDFS Client libraries to talk to a filesystem from an application. I've checked out Hadoop, made minor changes and built it with Maven, and tracked down the resulting artifacts in a target/ directory that I could deploy. Is this typically how a cloudera/hortonworks/mapr/etc dev works, or are the IDEs more common? I realize this sort of sounds like a dumb question, but I'm mostly curious what I might be missing out on if I stay away from anything other than vim, and not being entirely sure where maven might be caching jars that it uses to build, and how careful I have to be to ensure that my changes wind up in the right places without having to do a clean build every time. Thanks! -Erik -- Todd Lipcon Software Engineer, Cloudera
Re: Hadoop build slaves software
I'll install the right protoc and libstdc++ dev on asf009 as well. -Todd On Mon, Jan 7, 2013 at 9:57 AM, Andrew Wang andrew.w...@cloudera.com wrote: I think hadoop9 has a similar problem as hadoop8, based on a recent build. The javac output has a compile-proto error: https://builds.apache.org/job/PreCommit-HDFS-Build/3755/ https://builds.apache.org/job/PreCommit-HDFS-Build/3755/artifact/trunk/patchprocess/trunkJavacWarnings.txt On Sun, Jan 6, 2013 at 1:57 AM, Binglin Chang decst...@gmail.com wrote: HAServiceProtocol.proto:21:8: Option java_generic_services unknown. This is probably caused by a older version of protoc in build env. On Sun, Jan 6, 2013 at 2:12 PM, Giridharan Kesavan gkesa...@hortonworks.com wrote: by looking at the failure log : https://builds.apache.org/view/Hadoop/job/PreCommit-HADOOP-Build/1950/artifact/trunk/patchprocess/trunkJavacWarnings.txt build failed on [INFO] --- exec-maven-plugin:1.2:exec (compile-proto) @ hadoop-common --- HAServiceProtocol.proto:21:8: Option java_generic_services unknown. I'm not sure if this is something to do with the build env. -Giri On Sat, Jan 5, 2013 at 5:57 PM, Binglin Chang decst...@gmail.com wrote: I am not sure if this problem is solved, the build still failed in precommit-HADOOP https://builds.apache.org/view/Hadoop/job/PreCommit-HADOOP-Build/ On Sat, Jan 5, 2013 at 6:46 AM, Giridharan Kesavan gkesa...@hortonworks.com wrote: Marking the slave offline would do. I 've mared the hadoop8 slave offline, while I test it for builds and bring it back online later when its good. -Giri On Fri, Jan 4, 2013 at 2:26 PM, Todd Lipcon t...@cloudera.com wrote: Turns out I had to both kill -9 it and chmod 000 /home/jenkins/jenkins-slave in order to keep it from auto-respawning. Just a note so that once the toolchain is fixed, someone knows to re-chmod back to 755. -Todd On Fri, Jan 4, 2013 at 2:11 PM, Todd Lipcon t...@cloudera.com wrote: I'm going to kill -9 the jenkins slave on hadoop8 for now cuz it's causing havoc on the precommit builds. I can't see another way to administratively disable it from the Jenkins interface. Rajiv, Giri -- mind if I build/install protoc into /usr/local to match the other slaves? We can continue the conversation about provisioning after, but would like to unblock the builds in the meantime. As for CentOS vs Ubuntu, I've got no preference. RHEL6 is probably preferable since it's a more common install platform, anyway. But, we'll still need to have a custom toolchain for things like protoc 2.4 which don't have new enough versions in the package repos. -Todd On Fri, Jan 4, 2013 at 2:03 PM, Colin McCabe cmcc...@alumni.cmu.edu wrote: In addition to protoc, can someone please also install a 32-bit C++ compiler? The builds are all failing on this machine because of that. regards, Colin On Fri, Jan 4, 2013 at 11:37 AM, Giridharan Kesavan gkesa...@hortonworks.com wrote: When I configured the other machines I used the source to compile and install the protoc, as the 2.4.1 wasn't available in the ubuntu repo. BTW installed 2.4.1 on asf008. gkesavan@asf008:~$ protoc --version libprotoc 2.4.1 -Giri On Thu, Jan 3, 2013 at 11:24 PM, Todd Lipcon t...@cloudera.com wrote: Hey folks, It looks like hadoop8 has recently come back online as a build slave, but is failing all the builds because it has an ancient version of protobuf (2.2.0): todd@asf008:~$ protoc --version libprotoc 2.2.0 In contrast, other slaves have 2.4.1: todd@asf001:~$ protoc --version libprotoc 2.4.1 asf001 has the newer protoc in /usr/local/bin but asf008 does not. Does anyone know how software is meant to be deployed on these build slaves? I'm happy to download and install protobuf 2.4.1 into /usr/local on asf008 if manual installation is the name of the game, but it seems like we should be doing something a little more reproducible than one-off builds by rando developers to manage our toolchain on the Jenkins slaves. -Todd -- Todd Lipcon Software Engineer, Cloudera -- Todd Lipcon Software Engineer, Cloudera -- Todd Lipcon Software Engineer, Cloudera -- Todd Lipcon Software Engineer, Cloudera
Re: Hadoop build slaves software
OK. FYI, installed protoc on asf009, and the g++-4.4-multilib packages on both asf008 and asf009. Checked the hadoop pipes native build and it passes now. Fingers crossed... -Todd On Mon, Jan 7, 2013 at 3:35 PM, Todd Lipcon t...@cloudera.com wrote: I'll install the right protoc and libstdc++ dev on asf009 as well. -Todd On Mon, Jan 7, 2013 at 9:57 AM, Andrew Wang andrew.w...@cloudera.com wrote: I think hadoop9 has a similar problem as hadoop8, based on a recent build. The javac output has a compile-proto error: https://builds.apache.org/job/PreCommit-HDFS-Build/3755/ https://builds.apache.org/job/PreCommit-HDFS-Build/3755/artifact/trunk/patchprocess/trunkJavacWarnings.txt On Sun, Jan 6, 2013 at 1:57 AM, Binglin Chang decst...@gmail.com wrote: HAServiceProtocol.proto:21:8: Option java_generic_services unknown. This is probably caused by a older version of protoc in build env. On Sun, Jan 6, 2013 at 2:12 PM, Giridharan Kesavan gkesa...@hortonworks.com wrote: by looking at the failure log : https://builds.apache.org/view/Hadoop/job/PreCommit-HADOOP-Build/1950/artifact/trunk/patchprocess/trunkJavacWarnings.txt build failed on [INFO] --- exec-maven-plugin:1.2:exec (compile-proto) @ hadoop-common --- HAServiceProtocol.proto:21:8: Option java_generic_services unknown. I'm not sure if this is something to do with the build env. -Giri On Sat, Jan 5, 2013 at 5:57 PM, Binglin Chang decst...@gmail.com wrote: I am not sure if this problem is solved, the build still failed in precommit-HADOOP https://builds.apache.org/view/Hadoop/job/PreCommit-HADOOP-Build/ On Sat, Jan 5, 2013 at 6:46 AM, Giridharan Kesavan gkesa...@hortonworks.com wrote: Marking the slave offline would do. I 've mared the hadoop8 slave offline, while I test it for builds and bring it back online later when its good. -Giri On Fri, Jan 4, 2013 at 2:26 PM, Todd Lipcon t...@cloudera.com wrote: Turns out I had to both kill -9 it and chmod 000 /home/jenkins/jenkins-slave in order to keep it from auto-respawning. Just a note so that once the toolchain is fixed, someone knows to re-chmod back to 755. -Todd On Fri, Jan 4, 2013 at 2:11 PM, Todd Lipcon t...@cloudera.com wrote: I'm going to kill -9 the jenkins slave on hadoop8 for now cuz it's causing havoc on the precommit builds. I can't see another way to administratively disable it from the Jenkins interface. Rajiv, Giri -- mind if I build/install protoc into /usr/local to match the other slaves? We can continue the conversation about provisioning after, but would like to unblock the builds in the meantime. As for CentOS vs Ubuntu, I've got no preference. RHEL6 is probably preferable since it's a more common install platform, anyway. But, we'll still need to have a custom toolchain for things like protoc 2.4 which don't have new enough versions in the package repos. -Todd On Fri, Jan 4, 2013 at 2:03 PM, Colin McCabe cmcc...@alumni.cmu.edu wrote: In addition to protoc, can someone please also install a 32-bit C++ compiler? The builds are all failing on this machine because of that. regards, Colin On Fri, Jan 4, 2013 at 11:37 AM, Giridharan Kesavan gkesa...@hortonworks.com wrote: When I configured the other machines I used the source to compile and install the protoc, as the 2.4.1 wasn't available in the ubuntu repo. BTW installed 2.4.1 on asf008. gkesavan@asf008:~$ protoc --version libprotoc 2.4.1 -Giri On Thu, Jan 3, 2013 at 11:24 PM, Todd Lipcon t...@cloudera.com wrote: Hey folks, It looks like hadoop8 has recently come back online as a build slave, but is failing all the builds because it has an ancient version of protobuf (2.2.0): todd@asf008:~$ protoc --version libprotoc 2.2.0 In contrast, other slaves have 2.4.1: todd@asf001:~$ protoc --version libprotoc 2.4.1 asf001 has the newer protoc in /usr/local/bin but asf008 does not. Does anyone know how software is meant to be deployed on these build slaves? I'm happy to download and install protobuf 2.4.1 into /usr/local on asf008 if manual installation is the name of the game, but it seems like we should be doing something a little more reproducible than one-off builds by rando developers to manage our toolchain on the Jenkins slaves. -Todd -- Todd Lipcon Software Engineer, Cloudera -- Todd Lipcon Software Engineer, Cloudera -- Todd Lipcon Software Engineer, Cloudera
Re: Hadoop build slaves software
It was missing the c++ libraries though: https://builds.apache.org/job/PreCommit-HADOOP-Build/2002/artifact/trunk/patchprocess/patchJavacWarnings.txt -Todd On Mon, Jan 7, 2013 at 3:46 PM, Giridharan Kesavan gkesa...@hortonworks.com wrote: I did install protoc on hadoop9 and brought it back online after testing it couple of hours back. -Giri On Mon, Jan 7, 2013 at 3:35 PM, Todd Lipcon t...@cloudera.com wrote: I'll install the right protoc and libstdc++ dev on asf009 as well. -Todd On Mon, Jan 7, 2013 at 9:57 AM, Andrew Wang andrew.w...@cloudera.com wrote: I think hadoop9 has a similar problem as hadoop8, based on a recent build. The javac output has a compile-proto error: https://builds.apache.org/job/PreCommit-HDFS-Build/3755/ https://builds.apache.org/job/PreCommit-HDFS-Build/3755/artifact/trunk/patchprocess/trunkJavacWarnings.txt On Sun, Jan 6, 2013 at 1:57 AM, Binglin Chang decst...@gmail.com wrote: HAServiceProtocol.proto:21:8: Option java_generic_services unknown. This is probably caused by a older version of protoc in build env. On Sun, Jan 6, 2013 at 2:12 PM, Giridharan Kesavan gkesa...@hortonworks.com wrote: by looking at the failure log : https://builds.apache.org/view/Hadoop/job/PreCommit-HADOOP-Build/1950/artifact/trunk/patchprocess/trunkJavacWarnings.txt build failed on [INFO] --- exec-maven-plugin:1.2:exec (compile-proto) @ hadoop-common --- HAServiceProtocol.proto:21:8: Option java_generic_services unknown. I'm not sure if this is something to do with the build env. -Giri On Sat, Jan 5, 2013 at 5:57 PM, Binglin Chang decst...@gmail.com wrote: I am not sure if this problem is solved, the build still failed in precommit-HADOOP https://builds.apache.org/view/Hadoop/job/PreCommit-HADOOP-Build/ On Sat, Jan 5, 2013 at 6:46 AM, Giridharan Kesavan gkesa...@hortonworks.com wrote: Marking the slave offline would do. I 've mared the hadoop8 slave offline, while I test it for builds and bring it back online later when its good. -Giri On Fri, Jan 4, 2013 at 2:26 PM, Todd Lipcon t...@cloudera.com wrote: Turns out I had to both kill -9 it and chmod 000 /home/jenkins/jenkins-slave in order to keep it from auto-respawning. Just a note so that once the toolchain is fixed, someone knows to re-chmod back to 755. -Todd On Fri, Jan 4, 2013 at 2:11 PM, Todd Lipcon t...@cloudera.com wrote: I'm going to kill -9 the jenkins slave on hadoop8 for now cuz it's causing havoc on the precommit builds. I can't see another way to administratively disable it from the Jenkins interface. Rajiv, Giri -- mind if I build/install protoc into /usr/local to match the other slaves? We can continue the conversation about provisioning after, but would like to unblock the builds in the meantime. As for CentOS vs Ubuntu, I've got no preference. RHEL6 is probably preferable since it's a more common install platform, anyway. But, we'll still need to have a custom toolchain for things like protoc 2.4 which don't have new enough versions in the package repos. -Todd On Fri, Jan 4, 2013 at 2:03 PM, Colin McCabe cmcc...@alumni.cmu.edu wrote: In addition to protoc, can someone please also install a 32-bit C++ compiler? The builds are all failing on this machine because of that. regards, Colin On Fri, Jan 4, 2013 at 11:37 AM, Giridharan Kesavan gkesa...@hortonworks.com wrote: When I configured the other machines I used the source to compile and install the protoc, as the 2.4.1 wasn't available in the ubuntu repo. BTW installed 2.4.1 on asf008. gkesavan@asf008:~$ protoc --version libprotoc 2.4.1 -Giri On Thu, Jan 3, 2013 at 11:24 PM, Todd Lipcon t...@cloudera.com wrote: Hey folks, It looks like hadoop8 has recently come back online as a build slave, but is failing all the builds because it has an ancient version of protobuf (2.2.0): todd@asf008:~$ protoc --version libprotoc 2.2.0 In contrast, other slaves have 2.4.1: todd@asf001:~$ protoc --version libprotoc 2.4.1 asf001 has the newer protoc in /usr/local/bin but asf008 does not. Does anyone know how software is meant to be deployed on these build slaves? I'm happy to download and install protobuf 2.4.1 into /usr/local on asf008 if manual installation is the name of the game, but it seems like we should be doing something a little more reproducible than one-off
[jira] [Created] (HADOOP-9150) Unnecessary DNS resolution attempts for logical URIs
Todd Lipcon created HADOOP-9150: --- Summary: Unnecessary DNS resolution attempts for logical URIs Key: HADOOP-9150 URL: https://issues.apache.org/jira/browse/HADOOP-9150 Project: Hadoop Common Issue Type: Bug Components: ha Affects Versions: 2.0.2-alpha, 3.0.0 Reporter: Todd Lipcon Priority: Critical In the FileSystem code, we accidentally try to DNS-resolve the logical name before it is converted to an actual domain name. In some DNS setups, this can cause a big slowdown - eg in one misconfigured cluster we saw a 2-3x drop in terasort throughput, since every task wasted a lot of time waiting for slow not found responses from DNS. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
Unstuck QA bot
The QA bot was previously pointing at a JIRA filter which matched all Patch Available issues. That number has grown to 300 across the many projects that the QA bot works on. Unfortunately, that meant that a lot of the newer JIRAs (especially in alphabetically lower projects like HBASE, HADOOP, and GIRAPH) weren't getting picked up. I duped the filter, restricted it to those PA issues updated in the last 2 weeks, and changed the jenkins PreCommit-Admin job to point to the new one. It seems to have now unstuck itself and is processing a bunch of attachments that it had fallen behind on. Thanks -Todd -- Todd Lipcon Software Engineer, Cloudera
[jira] [Created] (HADOOP-9112) test-patch should -1 for @Tests without a timeout
Todd Lipcon created HADOOP-9112: --- Summary: test-patch should -1 for @Tests without a timeout Key: HADOOP-9112 URL: https://issues.apache.org/jira/browse/HADOOP-9112 Project: Hadoop Common Issue Type: Improvement Reporter: Todd Lipcon With our current test running infrastructure, if a test with no timeout set runs too long, it triggers a surefire-wide timeout, which for some reason doesn't show up as a failed test in the test-patch output. Given that, we should require that all tests have a timeout set, and have test-patch enforce this with a simple check -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-9106) Allow configuration of IPC connect timeout
Todd Lipcon created HADOOP-9106: --- Summary: Allow configuration of IPC connect timeout Key: HADOOP-9106 URL: https://issues.apache.org/jira/browse/HADOOP-9106 Project: Hadoop Common Issue Type: Improvement Components: ipc Affects Versions: 3.0.0 Reporter: Todd Lipcon Currently the connection timeout in Client.setupConnection() is hard coded to 20seconds. This is unreasonable in some scenarios, such as HA failover, if we want a faster failover time. We should allow this to be configured per-client. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
Re: Why don't the ipc Server use ArrayBlockingQueue for callQueue?
Hi Luoli Why would it be more efficient? When suggesting an improvement, it would be good to back it up with your reasoning. -Todd On Wed, Oct 31, 2012 at 1:19 AM, 罗李 luoli...@gmail.com wrote: hi everybody: I have a little question, why don't the ipc Server in hadoop use ArrayBlockingQueue for the callQueue but use LinkedBlockingQueue? Will it be more efficiently? thanks luoli -- Todd Lipcon Software Engineer, Cloudera
[jira] [Created] (HADOOP-8929) Add toString for SampleQuantiles
Todd Lipcon created HADOOP-8929: --- Summary: Add toString for SampleQuantiles Key: HADOOP-8929 URL: https://issues.apache.org/jira/browse/HADOOP-8929 Project: Hadoop Common Issue Type: Improvement Components: metrics Affects Versions: 2.0.2-alpha, 3.0.0 Reporter: Todd Lipcon The new SampleQuantiles class is useful in the context of benchmarks, but currently there is no way to print it out outside the context of a metrics sink. It would be nice to have a convenient way to stringify it for logging, etc. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-8905) Add metrics for HTTP Server
Todd Lipcon created HADOOP-8905: --- Summary: Add metrics for HTTP Server Key: HADOOP-8905 URL: https://issues.apache.org/jira/browse/HADOOP-8905 Project: Hadoop Common Issue Type: Improvement Components: metrics Affects Versions: 3.0.0 Reporter: Todd Lipcon Currently we don't expose any metrics about the HTTP server. It would be useful to be able to monitor the following: - Number of threads currently actively serving servlet requests - Total number of requests served - Perhaps break down time/count by endpoint (eg /jmx, /conf, various JSPs) This becomes more important as http-based protocols like webhdfs become more common -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-8889) Upgrade to Surefire 2.12.3
Todd Lipcon created HADOOP-8889: --- Summary: Upgrade to Surefire 2.12.3 Key: HADOOP-8889 URL: https://issues.apache.org/jira/browse/HADOOP-8889 Project: Hadoop Common Issue Type: Improvement Components: build, test Affects Versions: 3.0.0 Reporter: Todd Lipcon Assignee: Todd Lipcon Attachments: hadoop-8889.txt Surefire 2.12.3 has a couple improvements which are helpful for us. In particular, it fixes http://jira.codehaus.org/browse/SUREFIRE-817 which has been aggravating in the past. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-8894) GenericTestUtils.waitFor should dump thread stacks on timeout
Todd Lipcon created HADOOP-8894: --- Summary: GenericTestUtils.waitFor should dump thread stacks on timeout Key: HADOOP-8894 URL: https://issues.apache.org/jira/browse/HADOOP-8894 Project: Hadoop Common Issue Type: Improvement Reporter: Todd Lipcon Assignee: Todd Lipcon Many tests use this utility to wait for a condition to become true. In the event that it times out, we should dump all the thread stack traces, in case the timeout was due to a deadlock. This should make it easier to debug scenarios like HDFS-4001. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-8855) SSL-based image transfer does not work when Kerberos is disabled
Todd Lipcon created HADOOP-8855: --- Summary: SSL-based image transfer does not work when Kerberos is disabled Key: HADOOP-8855 URL: https://issues.apache.org/jira/browse/HADOOP-8855 Project: Hadoop Common Issue Type: Bug Components: security Affects Versions: 3.0.0, 2.0.2-alpha Reporter: Todd Lipcon Assignee: Todd Lipcon Priority: Minor In SecurityUtil.openSecureHttpConnection, we first check {{UserGroupInformation.isSecurityEnabled()}}. However, this only checks the kerberos config, which is independent of {{hadoop.ssl.enabled}}. Instead, we should check {{HttpConfig.isSecure()}}. Credit to Wing Yew Poon for discovering this bug -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
Re: Test timeouts
On Fri, Sep 14, 2012 at 11:23 AM, Aaron T. Myers a...@cloudera.com wrote: So, if you see a test case fail by reaching the Surefire fork timeout, please file a JIRA to add a JUnit timeout for that test. If when adding a test case you think that it might time out, please add a JUnit timeout. I'd go one step further: If you add any test that isn't a true unit test (ie it relies on any mulithreading, either explicitly or by using miniclusters, etc), you should add a timeout. Even if it has to be conservative (like 5 minutes on a test that you expect only runs 10 seconds), it seems better than waiting for it to timeout and then later having to go back and add one. -Todd -- Todd Lipcon Software Engineer, Cloudera
[jira] [Created] (HADOOP-8786) HttpServer continues to start even if AuthenticationFilter fails to init
Todd Lipcon created HADOOP-8786: --- Summary: HttpServer continues to start even if AuthenticationFilter fails to init Key: HADOOP-8786 URL: https://issues.apache.org/jira/browse/HADOOP-8786 Project: Hadoop Common Issue Type: Bug Affects Versions: 2.0.1-alpha, 1.2.0, 3.0.0 Reporter: Todd Lipcon Assignee: Todd Lipcon As seen in HDFS-3904, if the AuthenticationFilter fails to initialize, the web server will continue to start up. We need to check for context initialization errors after starting the server. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-8757) Metrics should disallow names with invalid characters
Todd Lipcon created HADOOP-8757: --- Summary: Metrics should disallow names with invalid characters Key: HADOOP-8757 URL: https://issues.apache.org/jira/browse/HADOOP-8757 Project: Hadoop Common Issue Type: Improvement Components: metrics Affects Versions: 2.0.0-alpha, 3.0.0 Reporter: Todd Lipcon Priority: Minor Just spent a couple hours trying to figure out why a metric I added didn't show up in JMX, only to eventually realize it was because I had a whitespace in the property name. This didn't cause any errors to be logged -- the metric just didn't show up in JMX. We should check that the name is valid and log an error, or replace invalid characters with something like an underscore. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Resolved] (HADOOP-8031) Configuration class fails to find embedded .jar resources; should use URL.openStream()
[ https://issues.apache.org/jira/browse/HADOOP-8031?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Todd Lipcon resolved HADOOP-8031. - Resolution: Fixed Re-resolving this since Ahmed is addressing my issue in HADOOP-8749 Configuration class fails to find embedded .jar resources; should use URL.openStream() -- Key: HADOOP-8031 URL: https://issues.apache.org/jira/browse/HADOOP-8031 Project: Hadoop Common Issue Type: Bug Components: conf Affects Versions: 2.0.0-alpha Reporter: Elias Ross Assignee: Elias Ross Fix For: 2.2.0-alpha Attachments: 0001-fix-HADOOP-7982-class-loader.patch, HADOOP-8031-part2.patch, HADOOP-8031.patch, hadoop-8031.txt While running a hadoop client within RHQ (monitoring software) using its classloader, I see this: 2012-02-07 09:15:25,313 INFO [ResourceContainer.invoker.daemon-2] (org.apache.hadoop.conf.Configuration)- parsing jar:file:/usr/local/rhq-agent/data/tmp/rhq-hadoop-plugin-4.3.0-SNAPSHOT.jar6856622641102893436.classloader/hadoop-core-0.20.2+737+1.jar7204287718482036191.tmp!/core-default.xml 2012-02-07 09:15:25,318 ERROR [InventoryManager.discovery-1] (rhq.core.pc.inventory.InventoryManager)- Failed to start component for Resource[id=16290, type=NameNode, key=NameNode:/usr/lib/hadoop-0.20, name=NameNode, parent=vg61l01ad-hadoop002.apple.com] from synchronized merge. org.rhq.core.clientapi.agent.PluginContainerException: Failed to start component for resource Resource[id=16290, type=NameNode, key=NameNode:/usr/lib/hadoop-0.20, name=NameNode, parent=vg61l01ad-hadoop002.apple.com]. Caused by: java.lang.RuntimeException: core-site.xml not found at org.apache.hadoop.conf.Configuration.loadResource(Configuration.java:1308) at org.apache.hadoop.conf.Configuration.loadResources(Configuration.java:1228) at org.apache.hadoop.conf.Configuration.getProps(Configuration.java:1169) at org.apache.hadoop.conf.Configuration.set(Configuration.java:438) This is because the URL jar:file:/usr/local/rhq-agent/data/tmp/rhq-hadoop-plugin-4.3.0-SNAPSHOT.jar6856622641102893436.classloader/hadoop-core-0.20.2+737+1.jar7204287718482036191.tmp!/core-default.xml cannot be found by DocumentBuilder (doesn't understand it). (Note: the logs are for an old version of Configuration class, but the new version has the same code.) The solution is to obtain the resource stream directly from the URL object itself. That is to say: {code} URL url = getResource((String)name); -if (url != null) { - if (!quiet) { -LOG.info(parsing + url); - } - doc = builder.parse(url.toString()); -} +doc = builder.parse(url.openStream()); {code} Note: I have a full patch pending approval at Apple for this change, including some cleanup. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Reopened] (HADOOP-8031) Configuration class fails to find embedded .jar resources; should use URL.openStream()
[ https://issues.apache.org/jira/browse/HADOOP-8031?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Todd Lipcon reopened HADOOP-8031: - I confirmed that reverting this patch locally restored the old behavior. If we can't maintain the old behavior, we should at least mark this as an incompatible change. But I bet it's doable to both fix it and have relative xincludes. Configuration class fails to find embedded .jar resources; should use URL.openStream() -- Key: HADOOP-8031 URL: https://issues.apache.org/jira/browse/HADOOP-8031 Project: Hadoop Common Issue Type: Bug Components: conf Affects Versions: 2.0.0-alpha Reporter: Elias Ross Assignee: Elias Ross Fix For: 2.2.0-alpha Attachments: 0001-fix-HADOOP-7982-class-loader.patch, HADOOP-8031.patch, hadoop-8031.txt While running a hadoop client within RHQ (monitoring software) using its classloader, I see this: 2012-02-07 09:15:25,313 INFO [ResourceContainer.invoker.daemon-2] (org.apache.hadoop.conf.Configuration)- parsing jar:file:/usr/local/rhq-agent/data/tmp/rhq-hadoop-plugin-4.3.0-SNAPSHOT.jar6856622641102893436.classloader/hadoop-core-0.20.2+737+1.jar7204287718482036191.tmp!/core-default.xml 2012-02-07 09:15:25,318 ERROR [InventoryManager.discovery-1] (rhq.core.pc.inventory.InventoryManager)- Failed to start component for Resource[id=16290, type=NameNode, key=NameNode:/usr/lib/hadoop-0.20, name=NameNode, parent=vg61l01ad-hadoop002.apple.com] from synchronized merge. org.rhq.core.clientapi.agent.PluginContainerException: Failed to start component for resource Resource[id=16290, type=NameNode, key=NameNode:/usr/lib/hadoop-0.20, name=NameNode, parent=vg61l01ad-hadoop002.apple.com]. Caused by: java.lang.RuntimeException: core-site.xml not found at org.apache.hadoop.conf.Configuration.loadResource(Configuration.java:1308) at org.apache.hadoop.conf.Configuration.loadResources(Configuration.java:1228) at org.apache.hadoop.conf.Configuration.getProps(Configuration.java:1169) at org.apache.hadoop.conf.Configuration.set(Configuration.java:438) This is because the URL jar:file:/usr/local/rhq-agent/data/tmp/rhq-hadoop-plugin-4.3.0-SNAPSHOT.jar6856622641102893436.classloader/hadoop-core-0.20.2+737+1.jar7204287718482036191.tmp!/core-default.xml cannot be found by DocumentBuilder (doesn't understand it). (Note: the logs are for an old version of Configuration class, but the new version has the same code.) The solution is to obtain the resource stream directly from the URL object itself. That is to say: {code} URL url = getResource((String)name); -if (url != null) { - if (!quiet) { -LOG.info(parsing + url); - } - doc = builder.parse(url.toString()); -} +doc = builder.parse(url.openStream()); {code} Note: I have a full patch pending approval at Apple for this change, including some cleanup. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-8624) ProtobufRpcEngine should log all RPCs if TRACE logging is enabled
Todd Lipcon created HADOOP-8624: --- Summary: ProtobufRpcEngine should log all RPCs if TRACE logging is enabled Key: HADOOP-8624 URL: https://issues.apache.org/jira/browse/HADOOP-8624 Project: Hadoop Common Issue Type: Improvement Components: ipc Affects Versions: 3.0.0, 2.2.0-alpha Reporter: Todd Lipcon Assignee: Todd Lipcon Priority: Minor Since all RPC requests/responses are now ProtoBufs, it's easy to add a TRACE level logging output for ProtobufRpcEngine that actually shows the full content of all calls. This is very handy especially when writing/debugging unit tests, but might also be useful to enable at runtime for short periods of time to debug certain production issues. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-8608) Add Configuration API for parsing time durations
Todd Lipcon created HADOOP-8608: --- Summary: Add Configuration API for parsing time durations Key: HADOOP-8608 URL: https://issues.apache.org/jira/browse/HADOOP-8608 Project: Hadoop Common Issue Type: Improvement Components: conf Affects Versions: 3.0.0 Reporter: Todd Lipcon Hadoop has a lot of configurations which specify durations or intervals of time. Unfortunately these different configurations have little consistency in units - eg some are in milliseconds, some in seconds, and some in minutes. This makes it difficult for users to configure, since they have to always refer back to docs to remember the unit for each property. The proposed solution is to add an API like {{Configuration.getTimeDuration}} which allows the user to specify the units with a prefix. For example, 10ms, 10s, 10m, 10h, or even 10d. For backwards-compatibility, if the user does not specify a unit, the API can specify the default unit, and warn the user that they should specify an explicit unit instead. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Resolved] (HADOOP-8557) Core Test failed in jekins for patch pre-commit
[ https://issues.apache.org/jira/browse/HADOOP-8557?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Todd Lipcon resolved HADOOP-8557. - Resolution: Duplicate Resolving as dup of HADOOP-8537 Core Test failed in jekins for patch pre-commit Key: HADOOP-8557 URL: https://issues.apache.org/jira/browse/HADOOP-8557 Project: Hadoop Common Issue Type: Bug Components: test Affects Versions: 2.0.0-alpha Reporter: Junping Du Priority: Blocker In jenkins PreCommit build history (https://builds.apache.org/job/PreCommit-HADOOP-Build/), following tests are failed for all recently patches (build-1164,1166,1168,1170): org.apache.hadoop.ha.TestZKFailoverController.testGracefulFailover org.apache.hadoop.ha.TestZKFailoverController.testOneOfEverything org.apache.hadoop.io.file.tfile.TestTFileByteArrays.testOneBlock org.apache.hadoop.io.file.tfile.TestTFileByteArrays.testOneBlockPlusOneEntry org.apache.hadoop.io.file.tfile.TestTFileByteArrays.testThreeBlocks org.apache.hadoop.io.file.tfile.TestTFileJClassComparatorByteArrays.testOneBlock org.apache.hadoop.io.file.tfile.TestTFileJClassComparatorByteArrays.testOneBlockPlusOneEntry org.apache.hadoop.io.file.tfile.TestTFileJClassComparatorByteArrays.testThreeBlocks -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Resolved] (HADOOP-8423) MapFile.Reader.get() crashes jvm or throws EOFException on Snappy or LZO block-compressed data
[ https://issues.apache.org/jira/browse/HADOOP-8423?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Todd Lipcon resolved HADOOP-8423. - Resolution: Fixed Fix Version/s: 1.2.0 Committed to branch-1 for 1.2. Thanks for backporting, Harsh. MapFile.Reader.get() crashes jvm or throws EOFException on Snappy or LZO block-compressed data -- Key: HADOOP-8423 URL: https://issues.apache.org/jira/browse/HADOOP-8423 Project: Hadoop Common Issue Type: Bug Components: io Affects Versions: 0.20.2 Environment: Linux 2.6.32.23-0.3-default #1 SMP 2010-10-07 14:57:45 +0200 x86_64 x86_64 x86_64 GNU/Linux Reporter: Jason B Assignee: Todd Lipcon Fix For: 1.2.0, 2.1.0-alpha Attachments: HADOOP-8423-branch-1.patch, HADOOP-8423-branch-1.patch, MapFileCodecTest.java, hadoop-8423.txt I am using Cloudera distribution cdh3u1. When trying to check native codecs for better decompression performance such as Snappy or LZO, I ran into issues with random access using MapFile.Reader.get(key, value) method. First call of MapFile.Reader.get() works but a second call fails. Also I am getting different exceptions depending on number of entries in a map file. With LzoCodec and 10 record file, jvm gets aborted. At the same time the DefaultCodec works fine for all cases, as well as record compression for the native codecs. I created a simple test program (attached) that creates map files locally with sizes of 10 and 100 records for three codecs: Default, Snappy, and LZO. (The test requires corresponding native library available) The summary of problems are given below: Map Size: 100 Compression: RECORD == DefaultCodec: OK SnappyCodec: OK LzoCodec: OK Map Size: 10 Compression: RECORD == DefaultCodec: OK SnappyCodec: OK LzoCodec: OK Map Size: 100 Compression: BLOCK DefaultCodec: OK SnappyCodec: java.io.EOFException at org.apache.hadoop.io.compress.BlockDecompressorStream.getCompressedData(BlockDecompressorStream.java:114) LzoCodec: java.io.EOFException at org.apache.hadoop.io.compress.BlockDecompressorStream.getCompressedData(BlockDecompressorStream.java:114) Map Size: 10 Compression: BLOCK == DefaultCodec: OK SnappyCodec: java.lang.NoClassDefFoundError: Ljava/lang/InternalError at org.apache.hadoop.io.compress.snappy.SnappyDecompressor.decompressBytesDirect(Native Method) LzoCodec: # # A fatal error has been detected by the Java Runtime Environment: # # SIGSEGV (0xb) at pc=0x2b068ffcbc00, pid=6385, tid=47304763508496 # # JRE version: 6.0_21-b07 # Java VM: Java HotSpot(TM) 64-Bit Server VM (17.0-b17 mixed mode linux-amd64 ) # Problematic frame: # C [liblzo2.so.2+0x13c00] lzo1x_decompress+0x1a0 # -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-8590) Backport HADOOP-7318 (MD5Hash factory should reset the digester it returns) to branch-1
Todd Lipcon created HADOOP-8590: --- Summary: Backport HADOOP-7318 (MD5Hash factory should reset the digester it returns) to branch-1 Key: HADOOP-8590 URL: https://issues.apache.org/jira/browse/HADOOP-8590 Project: Hadoop Common Issue Type: Bug Components: io Affects Versions: 1.0.3 Reporter: Todd Lipcon I ran into this bug on branch-1 today, it seems like we should backport it. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
Re: Jetty fixes for Hadoop
+1 from me too. We've had this in CDH since Sep '11 and been working much better than the stock 6.1.26. -Todd On Tue, Jul 10, 2012 at 3:14 PM, Owen O'Malley omal...@apache.org wrote: On Tue, Jul 10, 2012 at 2:59 PM, Thomas Graves tgra...@yahoo-inc.comwrote: I'm +1 for adding it. I'm +1 also. -- Owen -- Todd Lipcon Software Engineer, Cloudera
Re: PreCommit-Admin not running
I reported it to builds@ earlier this week, seems to be running sporadically now. We had some of our build slaves dead due to the leap second bug. I did the date trick to try to fix them, but maybe Hudson also needs a restart on the slaves? On Mon, Jul 2, 2012 at 7:31 PM, Jun Ping Du j...@vmware.com wrote: Move to dev alias, it seems to stop working since weekend. Thanks, Junping - Original Message - From: Kihwal Lee kih...@yahoo-inc.com To: gene...@hadoop.apache.org Sent: Tuesday, July 3, 2012 3:59:28 AM Subject: PreCommit-Admin not running It looks like the PreCommit-Admin build job is not running. Can anyone give it a gentle nudge? Kihwal -- Todd Lipcon Software Engineer, Cloudera
Scheduled jenkins builds not running?
Our Precommit build (https://builds.apache.org/job/PreCommit-Admin/) appears to have stopped running at 7/2 5PM. I've manually triggered it a couple times since, but otherwise not running. Looking at the config, it doesn't look like anyone explicitly disabled it. Any idea what might be up? #asfinfra directed me to mail builds@ -Todd -- Todd Lipcon Software Engineer, Cloudera
[jira] [Created] (HADOOP-8537) Two TFile tests failing recently
Todd Lipcon created HADOOP-8537: --- Summary: Two TFile tests failing recently Key: HADOOP-8537 URL: https://issues.apache.org/jira/browse/HADOOP-8537 Project: Hadoop Common Issue Type: Bug Components: io Affects Versions: 3.0.0 Reporter: Todd Lipcon TestTFileJClassComparatorByteArrays and TestTFileByteArrays are failing in some recent patch builds (seems to have started in the middle of May). These tests previously failed in HADOOP-7111 - perhaps something regressed there? -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Resolved] (HADOOP-8529) Error while formatting the namenode in hadoop single node setup in windows
[ https://issues.apache.org/jira/browse/HADOOP-8529?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Todd Lipcon resolved HADOOP-8529. - Resolution: Invalid Please inquire with the user mailing lists for questions like this. JIRA is meant for bug/task tracking. Error while formatting the namenode in hadoop single node setup in windows -- Key: HADOOP-8529 URL: https://issues.apache.org/jira/browse/HADOOP-8529 Project: Hadoop Common Issue Type: Task Components: conf Affects Versions: 1.0.3 Environment: Windows XP using Cygwin Reporter: Narayana Karteek Priority: Blocker Attachments: capture8.bmp Original Estimate: 5h Remaining Estimate: 5h Hi, I tried to configure hadoop 1.0.3 .I added all libs from share folder to lib directory.But still i get the error while formatting the namenode $ ./hadoop namenode -format java.lang.NoClassDefFoundError: Caused by: java.lang.ClassNotFoundException: at java.net.URLClassLoader$1.run(URLClassLoader.java:202) at java.security.AccessController.doPrivileged(Native Method) at java.net.URLClassLoader.findClass(URLClassLoader.java:190) at java.lang.ClassLoader.loadClass(ClassLoader.java:306) at sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java:301) at java.lang.ClassLoader.loadClass(ClassLoader.java:247) . Program will exit.in class: Exception in thread main -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-8497) Shell needs a way to list amount of physical consumed space in a directory
Todd Lipcon created HADOOP-8497: --- Summary: Shell needs a way to list amount of physical consumed space in a directory Key: HADOOP-8497 URL: https://issues.apache.org/jira/browse/HADOOP-8497 Project: Hadoop Common Issue Type: Improvement Components: fs Affects Versions: 2.0.0-alpha, 1.0.3, 3.0.0 Reporter: Todd Lipcon Assignee: Andy Isaacson Currently, there is no way to see the physical consumed space for a directory. du lists the logical (pre-replication) space, and fs -count only displays the consumed space when a quota is set. This makes it hard for administrators to set a quota on a directory, since they have no way to determine a reasonable value. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-8410) SPNEGO filter should have better error messages when not fully configured
Todd Lipcon created HADOOP-8410: --- Summary: SPNEGO filter should have better error messages when not fully configured Key: HADOOP-8410 URL: https://issues.apache.org/jira/browse/HADOOP-8410 Project: Hadoop Common Issue Type: Bug Components: security Affects Versions: 2.0.0 Reporter: Todd Lipcon Priority: Minor I upgraded to a build which includes SPNEGO, but neglected to configure dfs.web.authentication.kerberos.principal. This resulted in the following error: 12/05/18 14:46:20 INFO server.KerberosAuthenticationHandler: Login using keytab //home/todd/confs/conf.pseudo.security//hdfs.keytab, for principal ${dfs.web.authentication.kerberos.principal} 12/05/18 14:46:20 WARN mortbay.log: failed SpnegoFilter: javax.servlet.ServletException: javax.security.auth.login.LoginException: Unable to obtain password from user Instead, it should give an error that the principal needs to be configured. Even better would be if we could default to HTTP/_HOST@default realm -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-8404) etc/hadoop in binary tarball missing hadoop-env.sh
Todd Lipcon created HADOOP-8404: --- Summary: etc/hadoop in binary tarball missing hadoop-env.sh Key: HADOOP-8404 URL: https://issues.apache.org/jira/browse/HADOOP-8404 Project: Hadoop Common Issue Type: Bug Affects Versions: 2.0.0 Reporter: Todd Lipcon todd@todd-w510:~/releases/hadoop-2.0.0-alpha$ ls etc/hadoop/ core-site.xml hdfs-site.xmlhttpfs-signature.secret slaves yarn-env.sh hadoop-metrics2.properties httpfs-env.shhttpfs-site.xml ssl-client.xml.example yarn-site.xml hadoop-metrics.properties httpfs-log4j.properties log4j.properties ssl-server.xml.example -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-8405) ZKFC tests leak ZK instances
Todd Lipcon created HADOOP-8405: --- Summary: ZKFC tests leak ZK instances Key: HADOOP-8405 URL: https://issues.apache.org/jira/browse/HADOOP-8405 Project: Hadoop Common Issue Type: Bug Components: auto-failover, test Affects Versions: Auto Failover (HDFS-3042) Reporter: Todd Lipcon Assignee: Todd Lipcon Priority: Minor Attachments: hadoop-8405.txt The ZKFC code wasn't previously terminating the ZK connection in all cases where it should (eg after a failed startup or after formatting ZK). This didn't cause a problem for CLI usage, since the process exited afterwards, but caused the test results to get clouded with a lot of Reconecting to ZK messages, which make the logs hard to read. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Resolved] (HADOOP-8405) ZKFC tests leak ZK instances
[ https://issues.apache.org/jira/browse/HADOOP-8405?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Todd Lipcon resolved HADOOP-8405. - Resolution: Fixed Fix Version/s: Auto Failover (HDFS-3042) Hadoop Flags: Reviewed Committed to branch, thanks Eli. ZKFC tests leak ZK instances Key: HADOOP-8405 URL: https://issues.apache.org/jira/browse/HADOOP-8405 Project: Hadoop Common Issue Type: Bug Components: auto-failover, test Affects Versions: Auto Failover (HDFS-3042) Reporter: Todd Lipcon Assignee: Todd Lipcon Priority: Minor Fix For: Auto Failover (HDFS-3042) Attachments: hadoop-8405.txt The ZKFC code wasn't previously terminating the ZK connection in all cases where it should (eg after a failed startup or after formatting ZK). This didn't cause a problem for CLI usage, since the process exited afterwards, but caused the test results to get clouded with a lot of Reconecting to ZK messages, which make the logs hard to read. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-8406) CompressionCodecFactory.CODEC_PROVIDERS iteration is thread-unsafe
Todd Lipcon created HADOOP-8406: --- Summary: CompressionCodecFactory.CODEC_PROVIDERS iteration is thread-unsafe Key: HADOOP-8406 URL: https://issues.apache.org/jira/browse/HADOOP-8406 Project: Hadoop Common Issue Type: Bug Components: io Affects Versions: 2.0.0 Reporter: Todd Lipcon Assignee: Todd Lipcon CompressionCodecFactory defines CODEC_PROVIDERS as: {code} private static final ServiceLoaderCompressionCodec CODEC_PROVIDERS = ServiceLoader.load(CompressionCodec.class); {code} but this is a lazy collection which is thread-unsafe to iterate. We either need to synchronize when we iterate over it, or we need to materialize it during class-loading time by copying to a non-lazy collection -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Resolved] (HADOOP-8276) Auto-HA: add config for java options to pass to zkfc daemon
[ https://issues.apache.org/jira/browse/HADOOP-8276?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Todd Lipcon resolved HADOOP-8276. - Resolution: Fixed Fix Version/s: Auto Failover (HDFS-3042) Hadoop Flags: Reviewed Auto-HA: add config for java options to pass to zkfc daemon --- Key: HADOOP-8276 URL: https://issues.apache.org/jira/browse/HADOOP-8276 Project: Hadoop Common Issue Type: Improvement Components: scripts Affects Versions: Auto Failover (HDFS-3042) Reporter: Todd Lipcon Assignee: Todd Lipcon Priority: Minor Fix For: Auto Failover (HDFS-3042) Attachments: hadoop-8276.txt Currently the zkfc daemon is started without any ability to specify java options for it. We should be add a flag so heap size, etc can be specified. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-8397) NPE thrown when IPC layer gets an EOF reading a response
Todd Lipcon created HADOOP-8397: --- Summary: NPE thrown when IPC layer gets an EOF reading a response Key: HADOOP-8397 URL: https://issues.apache.org/jira/browse/HADOOP-8397 Project: Hadoop Common Issue Type: Bug Components: ipc Affects Versions: 2.0.0 Reporter: Todd Lipcon Priority: Critical When making a call on an IPC connection where the other end has shut down, I see the following exception: Caused by: java.lang.NullPointerException at org.apache.hadoop.ipc.Client$Connection.receiveResponse(Client.java:852) at org.apache.hadoop.ipc.Client$Connection.run(Client.java:781) from the lines: {code} RpcResponseHeaderProto response = RpcResponseHeaderProto.parseDelimitedFrom(in); int callId = response.getCallId(); {code} This is because parseDelimitedFrom() returns null in the case that the next thing to be read on the stream is an EOF. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
Re: Sailfish
/ Sailfish tries to improve Hadoop-performance, particularly for large-jobs which process TB's of data and run for hours. In building Sailfish, we modify how map-output is handled and transported from map-reduce. The project pages provide more information about the project. We are looking for colloborators who can help get some of the ideas into Apache Hadoop. A possible step forward could be to make shuffle phase of Hadoop pluggable. If you are interested in working with us, please get in touch with me. Sriram -- Best regards, - Andy Problems worthy of attack prove their worth by hitting back. - Piet Hein (via Tom White) -- Todd Lipcon Software Engineer, Cloudera
[jira] [Created] (HADOOP-8362) Improve exception message when Configuration.set() is called with a null key or value
Todd Lipcon created HADOOP-8362: --- Summary: Improve exception message when Configuration.set() is called with a null key or value Key: HADOOP-8362 URL: https://issues.apache.org/jira/browse/HADOOP-8362 Project: Hadoop Common Issue Type: Improvement Components: conf Affects Versions: 2.0.0 Reporter: Todd Lipcon Priority: Trivial Currently, calling Configuration.set(...) with a null value results in a NullPointerException within Properties.setProperty. We should check for null key/value and throw a better exception. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-8344) Improve test-patch to make it easier to find javadoc warnings
Todd Lipcon created HADOOP-8344: --- Summary: Improve test-patch to make it easier to find javadoc warnings Key: HADOOP-8344 URL: https://issues.apache.org/jira/browse/HADOOP-8344 Project: Hadoop Common Issue Type: Improvement Components: build, test Reporter: Todd Lipcon Priority: Minor Often I have to spend a lot of time digging through logs to find javadoc warnings as the result of a test-patch. Similar to the improvement made in HADOOP-8339, we should do the following: - test-patch should only run javadoc on modules that have changed - the exclusions OK_JAVADOC should be per-project rather than cross-project - rather than just have a number, we should check in the actual list of warnings to ignore and then fuzzy-match the patch warnings against the exclude list. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Resolved] (HADOOP-8279) Auto-HA: Allow manual failover to be invoked from zkfc.
[ https://issues.apache.org/jira/browse/HADOOP-8279?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Todd Lipcon resolved HADOOP-8279. - Resolution: Fixed Hadoop Flags: Reviewed Committed to branch, thanks Aaron. Auto-HA: Allow manual failover to be invoked from zkfc. --- Key: HADOOP-8279 URL: https://issues.apache.org/jira/browse/HADOOP-8279 Project: Hadoop Common Issue Type: Improvement Components: auto-failover, ha Affects Versions: Auto Failover (HDFS-3042) Reporter: Mingjie Lai Assignee: Todd Lipcon Fix For: Auto Failover (HDFS-3042) Attachments: hadoop-8279.txt, hadoop-8279.txt, hadoop-8279.txt, hadoop-8279.txt, hadoop-8279.txt HADOOP-8247 introduces a configure flag to prevent potential status inconsistency between zkfc and namenode, by making auto and manual failover mutually exclusive. However, as described in 2.7.2 section of design doc at HDFS-2185, we should allow manual and auto failover co-exist, by: - adding some rpc interfaces at zkfc - manual failover shall be triggered by haadmin, and handled by zkfc if auto failover is enabled. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-8340) SNAPSHOT build versions should compare as less than their eventual final release
Todd Lipcon created HADOOP-8340: --- Summary: SNAPSHOT build versions should compare as less than their eventual final release Key: HADOOP-8340 URL: https://issues.apache.org/jira/browse/HADOOP-8340 Project: Hadoop Common Issue Type: Improvement Components: util Affects Versions: 2.0.0 Reporter: Todd Lipcon Assignee: Todd Lipcon Priority: Minor We recently added a utility function to compare two version strings, based on splitting on '.'s and comparing each component. However, it considers a version like 2.0.0-SNAPSHOT as being greater than 2.0.0. This isn't right, since SNAPSHOT builds come before the final release. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-8315) Support SASL-authenticated ZooKeeper in ActiveStandbyElector
Todd Lipcon created HADOOP-8315: --- Summary: Support SASL-authenticated ZooKeeper in ActiveStandbyElector Key: HADOOP-8315 URL: https://issues.apache.org/jira/browse/HADOOP-8315 Project: Hadoop Common Issue Type: Improvement Components: auto-failover, ha Affects Versions: Auto Failover (HDFS-3042) Reporter: Todd Lipcon Currently, if you try to use SASL-authenticated ZK with the ActiveStandbyElector, you run into a couple issues: 1) We hit ZOOKEEPER-1437 - we need to wait until we see SaslAuthenticated before we can make any requests 2) We currently throw a fatalError when we see the SaslAuthenticated callback on the connection watcher We need to wait for ZK-1437 upstream, and then upgrade to the fixed version for #1. For #2 we just need to add a case there and ignore it. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Resolved] (HADOOP-8306) ZKFC: improve error message when ZK is not running
[ https://issues.apache.org/jira/browse/HADOOP-8306?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Todd Lipcon resolved HADOOP-8306. - Resolution: Fixed Fix Version/s: Auto Failover (HDFS-3042) Hadoop Flags: Reviewed Committed to branch, thanks Eli ZKFC: improve error message when ZK is not running -- Key: HADOOP-8306 URL: https://issues.apache.org/jira/browse/HADOOP-8306 Project: Hadoop Common Issue Type: Improvement Components: auto-failover, ha Affects Versions: Auto Failover (HDFS-3042) Reporter: Todd Lipcon Assignee: Todd Lipcon Priority: Minor Fix For: Auto Failover (HDFS-3042) Attachments: hadoop-8306.txt Currently if you start the ZKFC without starting ZK, you get an ugly stack trace. We should improve the error message and give it a unique exit code. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-8306) ZKFC: improve error message when ZK is not running
Todd Lipcon created HADOOP-8306: --- Summary: ZKFC: improve error message when ZK is not running Key: HADOOP-8306 URL: https://issues.apache.org/jira/browse/HADOOP-8306 Project: Hadoop Common Issue Type: Improvement Components: auto-failover, ha Affects Versions: Auto Failover (HDFS-3042) Reporter: Todd Lipcon Assignee: Todd Lipcon Priority: Minor Currently if you start the ZKFC without starting ZK, you get an ugly stack trace. We should improve the error message and give it a unique exit code. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-8295) ToolRunner.confirmPrompt spins if stdin goes away
ToolRunner.confirmPrompt spins if stdin goes away - Key: HADOOP-8295 URL: https://issues.apache.org/jira/browse/HADOOP-8295 Project: Hadoop Common Issue Type: Bug Affects Versions: 2.0.0 Reporter: Todd Lipcon Priority: Minor Currently, ToolRunner.confirmPrompt treats a -1 response to its System.in.read() call the same as \n or \r. So, if stdin goes away, it can cause the JVM to spin in a loop. Instead, it should treat a -1 result as a n response. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Resolved] (HADOOP-8258) Add interfaces for compression codecs to use direct byte buffers
[ https://issues.apache.org/jira/browse/HADOOP-8258?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Todd Lipcon resolved HADOOP-8258. - Resolution: Duplicate Resolving as dup - please see HADOOP-8148. Add interfaces for compression codecs to use direct byte buffers Key: HADOOP-8258 URL: https://issues.apache.org/jira/browse/HADOOP-8258 Project: Hadoop Common Issue Type: New Feature Components: io, native, performance Affects Versions: 3.0.0 Reporter: Todd Lipcon Currently, the codec interface only provides input/output functions based on byte arrays. Given that most of the codecs are implemented in native code, this necessitates two extra copies - one to copy the input data to a direct buffer, and one to copy the output data back to a byte array. We should add interfaces to Decompressor/Compressor that can work directly with direct byte buffers to avoid these copies. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-8276) Auto-HA: add config for java options to pass to zkfc daemon
Auto-HA: add config for java options to pass to zkfc daemon --- Key: HADOOP-8276 URL: https://issues.apache.org/jira/browse/HADOOP-8276 Project: Hadoop Common Issue Type: Improvement Components: scripts Affects Versions: Auto Failover (HDFS-3042) Reporter: Todd Lipcon Assignee: Todd Lipcon Priority: Minor Currently the zkfc daemon is started without any ability to specify java options for it. We should be add a flag so heap size, etc can be specified. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-8277) Auto-HA: add basic HTTP interface to ZKFC
Auto-HA: add basic HTTP interface to ZKFC - Key: HADOOP-8277 URL: https://issues.apache.org/jira/browse/HADOOP-8277 Project: Hadoop Common Issue Type: Improvement Components: ha Affects Versions: Auto Failover (HDFS-3042) Reporter: Todd Lipcon Assignee: Todd Lipcon Currently the ZKFC exposes no interfaces at all. It would be useful to add a very basic web interface, which shows its current status. This would also expose the usual /jmx, /conf, etc servlets for easier debugging and monitoring. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Resolved] (HADOOP-8272) BytesWritable length problem
[ https://issues.apache.org/jira/browse/HADOOP-8272?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Todd Lipcon resolved HADOOP-8272. - Resolution: Invalid This is not a bug - this is the expected behavior of getBytes(). Please refer to the javadoc: {code} /** * Get the data backing the BytesWritable. Please use {@link #copyBytes()} * if you need the returned array to be precisely the length of the data. * @return The data is only valid between 0 and getLength() - 1. */ {code} BytesWritable length problem Key: HADOOP-8272 URL: https://issues.apache.org/jira/browse/HADOOP-8272 Project: Hadoop Common Issue Type: Bug Components: io Affects Versions: 0.20.205.0 Reporter: Simon Gilliot Labels: hadoop Original Estimate: 1h Remaining Estimate: 1h I tried to create my own Writable which contains a BytesWritable. In my conctructor, I tried to create an empty BytesWritable : BytesWritable key = new BytesWritable(); Next, in my readFields, I did : key.readFields(in); LOG.debug(Bytes.toString(key.getBytes())); The key contains much more bytes than I had wrote. In fact, if my BytesWritable contains 100 bytes, I thing that the readFields() of BytesWritable call : * setSize(0) (which seems useless since the values in the old range are preserved and any new values are undefined). * setSize(100) which extends the bytes array (by setCapacity) to 1.5 * the size (so 150) without initalizing it * readFully(bytes, 0, 100) which fill the bytes array from '0' to '100' offsets. And when I call getBytes() on it, the bytes array of 150 bytes is returned without any control. That seems possible that the same problem happens in other conditions, when we increase ths bytes array size. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Resolved] (HADOOP-8247) Auto-HA: add a config to enable auto-HA, which disables manual FC
[ https://issues.apache.org/jira/browse/HADOOP-8247?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Todd Lipcon resolved HADOOP-8247. - Resolution: Fixed Fix Version/s: Auto Failover (HDFS-3042) Hadoop Flags: Reviewed Committed to branch, thx for reviews all. Auto-HA: add a config to enable auto-HA, which disables manual FC - Key: HADOOP-8247 URL: https://issues.apache.org/jira/browse/HADOOP-8247 Project: Hadoop Common Issue Type: Improvement Components: auto-failover, ha Affects Versions: Auto Failover (HDFS-3042) Reporter: Todd Lipcon Assignee: Todd Lipcon Fix For: Auto Failover (HDFS-3042) Attachments: hadoop-8247.txt, hadoop-8247.txt, hadoop-8247.txt, hadoop-8247.txt, hadoop-8247.txt, hadoop-8247.txt Currently, if automatic failover is set up and running, and the user uses the haadmin -failover command, he or she can end up putting the system in an inconsistent state, where the state in ZK disagrees with the actual state of the world. To fix this, we should add a config flag which is used to enable auto-HA. When this flag is set, we should disallow use of the haadmin command to initiate failovers. We should refuse to run ZKFCs when the flag is not set. Of course, this flag should be scoped by nameservice. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-8263) Stringification of IPC calls not useful
Stringification of IPC calls not useful --- Key: HADOOP-8263 URL: https://issues.apache.org/jira/browse/HADOOP-8263 Project: Hadoop Common Issue Type: Bug Components: ipc Affects Versions: 2.0.0 Reporter: Todd Lipcon Assignee: Todd Lipcon Priority: Minor Since the Protobufification of Hadoop, the log messages on IPC exceptions on the server side now read like: 12/04/09 16:04:06 INFO ipc.Server: IPC Server handler 9 on 8021, call org.apache.hadoop.ipc.ProtobufRpcEngine$RpcRequestWritable@7087e9bf from 127.0.0.1:47989: error: org.apache.hadoop.ipc.StandbyException: Operation category READ is not supported in state standby The call should instead stringify the method name and the request protobuf (perhaps abbreviated if it is longer than a few hundred chars) -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Resolved] (HADOOP-8262) Between mapper and reducer, Hadoop inserts spaces into my string
[ https://issues.apache.org/jira/browse/HADOOP-8262?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Todd Lipcon resolved HADOOP-8262. - Resolution: Invalid This sounds like a bug in your code, or a misunderstanding, rather than a bug in MR. Please feel free to contact the mapreduce-user list for help. Between mapper and reducer, Hadoop inserts spaces into my string Key: HADOOP-8262 URL: https://issues.apache.org/jira/browse/HADOOP-8262 Project: Hadoop Common Issue Type: Bug Components: io Affects Versions: 0.20.0 Environment: Eclipse plugin, Windows Reporter: Adriana Sbircea In the mapper i send as key a number, and as value another number which has more than one digit, but i send them as Text objects. In my reducer all the values for a key have spaces between every digit of a value. I can't do my task because of this problem. I don't use combiners or something else. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Resolved] (HADOOP-8260) Auto-HA: Replace ClientBaseWithFixes with our own modified copy of the class
[ https://issues.apache.org/jira/browse/HADOOP-8260?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Todd Lipcon resolved HADOOP-8260. - Resolution: Fixed Fix Version/s: Auto Failover (HDFS-3042) Hadoop Flags: Reviewed Committed to branch. I ran all of the tests which inherit from this class before committing. Auto-HA: Replace ClientBaseWithFixes with our own modified copy of the class Key: HADOOP-8260 URL: https://issues.apache.org/jira/browse/HADOOP-8260 Project: Hadoop Common Issue Type: Test Components: auto-failover, test Affects Versions: Auto Failover (HDFS-3042) Reporter: Todd Lipcon Assignee: Todd Lipcon Priority: Minor Fix For: Auto Failover (HDFS-3042) Attachments: hadoop-8260.txt The class ClientBaseWithFixes is an attempt to add some workaround code to avoid spurious failures due to ZOOKEEPER-1438. But, even after making those workarounds, I've seen a few Jenkins failures due to that issue. Until ZK fixes this issue, I'd like to just copy the test infrastructure into our own code, and remove the problematic JMXEnv verifications. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-8257) Auto-HA: TestZKFailoverControllerStress occasionally fails with Mockito error
Auto-HA: TestZKFailoverControllerStress occasionally fails with Mockito error - Key: HADOOP-8257 URL: https://issues.apache.org/jira/browse/HADOOP-8257 Project: Hadoop Common Issue Type: Bug Components: auto-failover, test Affects Versions: Auto Failover (HDFS-3042) Reporter: Todd Lipcon Assignee: Todd Lipcon Priority: Trivial Attachments: hadoop-8257.txt Once in a while I've seen the following in TestZKFailoverControllerStress: Unfinished stubbing detected here: - at org.apache.hadoop.ha.TestZKFailoverControllerStress.testRandomHealthAndDisconnects(TestZKFailoverControllerStress.java:118) E.g. thenReturn() may be missing This is because we set up the mock answers _after_ starting the ZKFCs. So if the ZKFC calls the mock object while it's in the middle of the setup, this exception occurs. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Resolved] (HADOOP-8257) Auto-HA: TestZKFailoverControllerStress occasionally fails with Mockito error
[ https://issues.apache.org/jira/browse/HADOOP-8257?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Todd Lipcon resolved HADOOP-8257. - Resolution: Fixed Fix Version/s: Auto Failover (HDFS-3042) Hadoop Flags: Reviewed Auto-HA: TestZKFailoverControllerStress occasionally fails with Mockito error - Key: HADOOP-8257 URL: https://issues.apache.org/jira/browse/HADOOP-8257 Project: Hadoop Common Issue Type: Bug Components: auto-failover, test Affects Versions: Auto Failover (HDFS-3042) Reporter: Todd Lipcon Assignee: Todd Lipcon Priority: Trivial Fix For: Auto Failover (HDFS-3042) Attachments: hadoop-8257.txt Once in a while I've seen the following in TestZKFailoverControllerStress: Unfinished stubbing detected here: - at org.apache.hadoop.ha.TestZKFailoverControllerStress.testRandomHealthAndDisconnects(TestZKFailoverControllerStress.java:118) E.g. thenReturn() may be missing This is because we set up the mock answers _after_ starting the ZKFCs. So if the ZKFC calls the mock object while it's in the middle of the setup, this exception occurs. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-8258) Add interfaces for compression codecs to use direct byte buffers
Add interfaces for compression codecs to use direct byte buffers Key: HADOOP-8258 URL: https://issues.apache.org/jira/browse/HADOOP-8258 Project: Hadoop Common Issue Type: New Feature Components: io, native Affects Versions: 3.0.0 Reporter: Todd Lipcon Currently, the codec interface only provides input/output functions based on byte arrays. Given that most of the codecs are implemented in native code, this necessitates two extra copies - one to copy the input data to a direct buffer, and one to copy the output data back to a byte array. We should add interfaces to Decompressor/Compressor that can work directly with direct byte buffers to avoid these copies. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-8260) Auto-HA: Replace ClientBaseWithFixes with our own modified copy of the class
Auto-HA: Replace ClientBaseWithFixes with our own modified copy of the class Key: HADOOP-8260 URL: https://issues.apache.org/jira/browse/HADOOP-8260 Project: Hadoop Common Issue Type: Test Components: auto-failover, test Affects Versions: Auto Failover (HDFS-3042) Reporter: Todd Lipcon Assignee: Todd Lipcon Priority: Minor Attachments: hadoop-8260.txt The class ClientBaseWithFixes is an attempt to add some workaround code to avoid spurious failures due to ZOOKEEPER-1438. But, even after making those workarounds, I've seen a few Jenkins failures due to that issue. Until ZK fixes this issue, I'd like to just copy the test infrastructure into our own code, and remove the problematic JMXEnv verifications. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Resolved] (HADOOP-8259) Auto-HA: Replace ClientBaseWithFixes with our own modified copy of the class
[ https://issues.apache.org/jira/browse/HADOOP-8259?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Todd Lipcon resolved HADOOP-8259. - Resolution: Duplicate JIRA being funky today... clicked create once but got two copies. Dup of HADOOP-8260 Auto-HA: Replace ClientBaseWithFixes with our own modified copy of the class Key: HADOOP-8259 URL: https://issues.apache.org/jira/browse/HADOOP-8259 Project: Hadoop Common Issue Type: Test Components: auto-failover, test Affects Versions: Auto Failover (HDFS-3042) Reporter: Todd Lipcon Assignee: Todd Lipcon Priority: Minor The class ClientBaseWithFixes is an attempt to add some workaround code to avoid spurious failures due to ZOOKEEPER-1438. But, even after making those workarounds, I've seen a few Jenkins failures due to that issue. Until ZK fixes this issue, I'd like to just copy the test infrastructure into our own code, and remove the problematic JMXEnv verifications. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (HADOOP-8251) SecurityUtil.fetchServiceTicket broken after HADOOP-6941
SecurityUtil.fetchServiceTicket broken after HADOOP-6941 Key: HADOOP-8251 URL: https://issues.apache.org/jira/browse/HADOOP-8251 Project: Hadoop Common Issue Type: Bug Components: security Affects Versions: 1.1.0, 2.0.0 Reporter: Todd Lipcon Assignee: Todd Lipcon Priority: Blocker Attachments: hadoop-8251.txt HADOOP-6941 replaced direct references to some classes with reflective access so as to support other JDKs. Unfortunately there was a mistake in the name of the Krb5Util class, which broke fetchServiceTicket. This manifests itself as the inability to run checkpoints or other krb5-SSL HTTP-based transfers: java.lang.ClassNotFoundException: sun.security.jgss.krb5 -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira