Re: Errors while starting James 3.5 (Sieve)
Yea, it's me again, sorry for spamming the list. Somehow I think I'm more the forum/board guy, anyway ... As I said: it's somehow random. So just to see if I can get it build when docker and docker-compose is installed I fired up a vm (same environment as root) - and it failed almost random at [INFO] Apache James :: Mailbox :: HBase ... FAILURE Here's the log: Running org.apache.james.mailbox.hbase.HBaseMailboxManagerTest Tests run: 1, Failures: 0, Errors: 1, Skipped: 0, Time elapsed: 0.006 sec <<< FAILURE! - in org.apache.james.mailbox.hbase.HBaseMailboxManagerTest initializationError(org.apache.james.mailbox.hbase.HBaseMailboxManagerTest) Time elapsed: 0.003 sec <<< ERROR! java.lang.NoClassDefFoundError: Could not initialize class org.apache.james.mailbox.hbase.HBaseMailboxSessionMapperFactoryTest Running org.apache.james.mailbox.hbase.HBaseUtilsTest SLF4J: Class path contains multiple SLF4J bindings. SLF4J: Found binding in [jar:file:/home/vm/.m2/repository/org/slf4j/slf4j-simple/1.7.7/slf4j-simple-1.7.7.jar!/org/slf4j/impl/StaticLoggerBinder.class] SLF4J: Found binding in [jar:file:/home/vm/.m2/repository/org/slf4j/slf4j-log4j12/1.4.3/slf4j-log4j12-1.4.3.jar!/org/slf4j/impl/StaticLoggerBinder.class] SLF4J: See http://www.slf4j.org/codes.html#multiple_bindings for an explanation. SLF4J: Actual binding is of type [org.slf4j.impl.SimpleLoggerFactory] subscription toPut flagsToPut mailboxToPut getRowKey and UUIDFromRowKey Tests run: 5, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.079 sec - in org.apache.james.mailbox.hbase.HBaseUtilsTest Results : Tests in error: HBaseMailboxManagerTest.initializationError » NoClassDefFound Could not initia... HBaseMailboxSessionMapperFactoryTest.testCreateMailboxMapper » NoClassDefFound HBaseMailboxSessionMapperFactoryTest.testCreateMessageMapper » NoClassDefFound HBaseMailboxSessionMapperFactoryTest.testCreateSubscriptionMapper » NoClassDefFound HBaseMailboxSessionMapperFactoryTest.testGetModSeqProvider » NoClassDefFound C... HBaseMailboxSessionMapperFactoryTest.testGetUidProvider » ExceptionInInitializer HBaseMailboxMapperTest.testMailboxMapperScenario » ExceptionInInitializer HBaseMailboxMessageMapperTest.testMessageMapperScenario » ExceptionInInitializer HBaseUidAndModSeqProviderTest.testHighestModSeq » NoClassDefFound Could not in... HBaseUidAndModSeqProviderTest.testLastUid » ExceptionInInitializer HBaseUidAndModSeqProviderTest.testNextModSeq » NoClassDefFound Could not initi... HBaseUidAndModSeqProviderTest.testNextUid » NoClassDefFound Could not initiali... HBaseSubscriptionMapperTest.testDelete » NoClassDefFound Could not initialize ... HBaseSubscriptionMapperTest.testFindMailboxSubscriptionForUser » NoClassDefFound HBaseSubscriptionMapperTest.testFindSubscriptionsForUser » NoClassDefFound Cou... HBaseSubscriptionMapperTest.testSave » ExceptionInInitializer Tests run: 32, Failures: 0, Errors: 16, Skipped: 0 So I just retried it - but at least this time it failed on the same step. So just to be sure - clean wipe and another retry - same result. Ok, so what ever is wrong here - the only difference are the additional packages noted as requirement. Same system - and different result on 2nd retry to reproduce last error - and something different just because it's (almost) same system running in a VM? Either it's something native so closely hardware dependent it really differes on such a low level java isn't supposed to be there - or I'm too stupid to simply compile some java source. So, any source of "nightly builds" of current git repo? Matt Am 19.01.2017 um 23:01 schrieb Mathias Rodenberg: Hello Benoit, thanks for your reply. Here is my current build environment running on my root: OS: OpenSuSE Tumbleweed (based on 13.2 OVH) JDK: OpenJDK 8u111 x64 git: 2.11.0 maven: 3.3.9 repo: git://git.apache.org/james-project.git cloned on 18.01.17 ~ 19:30 GMT+1 / re-cloned 19.01.17 ~ 19:30 GMT+1 command: mvn clean package -Pwith-assembly Yesterday, as I wrote my mail, the build stuck somewhere on cassandra (can't remember wich - but I guess it was the first line where cassandra is noted as it failed pretty fast). I tried it again today to replicate the error I got and it ran through almost to the end and failed somewhere on mailet (strange - if a build fails it should fail on the same point on a 2nd try). So I cleared my local cache (removed .m2 and james-project) and re-cloned the current commit. Kept failing on the same point: >... Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 5.972 sec - in org.apache.james.transport.mailets.StripAttachmentTest Results : Tests in error: AmqpForwardAttachmentTest.:87 » IllegalState Could not find a valid Dock... Tests run: 9, Failures: 0, Errors: 1, Skipped: 0 >... [INFO] Apache James :: Server :: Mailets Integration Testing FAILURE [ 49.600 s] &
Re: Errors while starting James 3.5 (Sieve)
Hello Benoit, thanks for your reply. Here is my current build environment running on my root: OS: OpenSuSE Tumbleweed (based on 13.2 OVH) JDK: OpenJDK 8u111 x64 git: 2.11.0 maven: 3.3.9 repo: git://git.apache.org/james-project.git cloned on 18.01.17 ~ 19:30 GMT+1 / re-cloned 19.01.17 ~ 19:30 GMT+1 command: mvn clean package -Pwith-assembly Yesterday, as I wrote my mail, the build stuck somewhere on cassandra (can't remember wich - but I guess it was the first line where cassandra is noted as it failed pretty fast). I tried it again today to replicate the error I got and it ran through almost to the end and failed somewhere on mailet (strange - if a build fails it should fail on the same point on a 2nd try). So I cleared my local cache (removed .m2 and james-project) and re-cloned the current commit. Kept failing on the same point: >... Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 5.972 sec - in org.apache.james.transport.mailets.StripAttachmentTest Results : Tests in error: AmqpForwardAttachmentTest.:87 » IllegalState Could not find a valid Dock... Tests run: 9, Failures: 0, Errors: 1, Skipped: 0 >... [INFO] Apache James :: Server :: Mailets Integration Testing FAILURE [ 49.600 s] >... [ERROR] Failed to execute goal org.apache.maven.plugins:maven-surefire-plugin:2.19:test (default-test) on project james-server-mailets-integration-testing: There are test failures. [ERROR] [ERROR] Please refer to /home/cryptearth/james-project/server/mailet/integration-testing/target/surefire-reports for the individual test results. [ERROR] -> [Help 1] This is the log output: cryptearth@root1:~/james-project> cat server/mailet/integration-testing/target/surefire-reports/org.apache.james.transport.mailets.AmqpForwardAttachmentTest.txt --- Test set: org.apache.james.transport.mailets.AmqpForwardAttachmentTest --- Tests run: 1, Failures: 0, Errors: 1, Skipped: 0, Time elapsed: 10.79 sec <<< FAILURE! - in org.apache.james.transport.mailets.AmqpForwardAttachmentTest stripAttachmentShouldPutAttachmentsInMailAttributeWhenConfiguredForIt(org.apache.james.transport.mailets.AmqpForwardAttachmentTest) Time elapsed: 0.022 sec <<< ERROR! java.lang.IllegalStateException: Could not find a valid Docker environment. Please see logs and check configuration at org.apache.james.transport.mailets.AmqpForwardAttachmentTest.(AmqpForwardAttachmentTest.java:87) Well, as I'm reading the Exception, no, I don't have a Docker-environment set-up, but as I built my pack (24.11.16) it wasn't required and I got built it successfull. As the Exception states "valid Docker environment" I guess it's not as simple as just installing docker via package manager but you have to set it up running for the test to succeed, or to get to current built up and running. That's it for now from my side. Hope I could provide any help. As always, greetings from Germany, Matt Am 19.01.2017 um 03:09 schrieb Benoit Tellier: Hello Matt, Hello Christian, Every time we merge a pull request, we have a working build (both compilation and unit tests) as our CI run against GitHub pull requests. So your issues are not expected. Could you provide us the maven command you issue, the git project you are trying to build, your JDK version, maven version, as well as the error, and the exact failing maven project ? Best regards, Benoit Le 19/01/2017 à 00:42, cryptearth a écrit : Well, unfortunately I wasn't able to compile the current commit either. It fails on cassandra. As I said: wait a day or two for the next commit, re-clone the repo and try it again. It should work on at least one of the next 5 - 10 commits to get a working one. If you want I can do the compilation for you and can provide the built package, but be aware I would only deliever it as is w/o taking any responsebility for issues you might encounter when using the package compiled by me. If I can get a working build (I don't have the package I built myself handy) I can provide you with a link (would be send directly to you instead of the list). Greetings from Germany, Matt Christian Fischer schrieb Hey there, I tried to build the current beta6 but i did not have sucess I do not know how to do it correctly (building it), and since Google delived me nothing useful (maybe i am searching with wrong arguments ), i would like to ask you, if you have something like an instruction for me ?? Thanks 2017-01-03 5:50 GMT+01:00 Mathias Rodenberg <cryptea...@cryptearth.de>: Hey Christian, your log shows not only a fatal Exception regarding the Sieve class, but also some other serious Exceptions that some configs or classes are corrupt or missing. As this is beta5 - I'm guessing you just downloaded the binary file
Re: Errors while starting James 3.5 (Sieve)
Hey Christian, your log shows not only a fatal Exception regarding the Sieve class, but also some other serious Exceptions that some configs or classes are corrupt or missing. As this is beta5 - I'm guessing you just downloaded the binary files and tried to run them. Sadly, James3 isn't able to run out-of-the-box that easy and require some tweaking with the tounge at the right angle. I would recommend you to clone current beta6 git repo and built it from source. On a Linux system this is very easy by simply installing maven, git and java-devel packages, this should be easy on a Windows system too. Be aware: the current james-repo is an active development repo, so it's possible to clone a commit wich can't be build (happend to my several times). If so - just wait a day or two and re-clone the newest commit again - it then should be compileable (or repeat if needed until you get a clean build). After unpacking the fresh build there are only template configs, but some of them need to be activated for James to successfully start up. The startup sequence will tell you wich config-file can't be loaded - simply switch to config-dir, enable the needed config, and re-try until james successfully completely starts up for the first time. You then should be able to access it remotely with the james-cli config-tool to add your domain and users. For security reasons the first config you should alter before enabling it is smtpserver.xml: on line 91 switch the tag to "true" to prevent your mail-server being an open relay to the world. All the other settings should be fine as long as you need them to be secific values. Hope it helps, oh, and a happy new year to all subscribers yours, Matt Am 01.01.2017 um 22:43 schrieb Christian Fischer: Hey there, i have some errors while starting James ... I hope somebody can help me. Thanks in advantage ! INFO 22:41:18,520 | org.apache.james.container.spring.context.JamesServerApplicationContext | Refreshing org.apache.james.container.spring.context.JamesServerApplicationContext@31a5c39e: startup date [Sun Jan 01 22:41:18 CET 2017]; root of context hierarchy DEBUG 22:41:24,319 | org.apache.james.container.spring.context.JamesServerApplicationContext | Bean factory for org.apache.james.container.spring.context.JamesServerApplicationContext@31a5c39e: org.springframework.beans.factory.support.DefaultListableBeanFactory@54d1608f: defining beans