About burakongun

burakongun · ‎05-20-2019

Hi, have you been able to solve this issue? I am having the same problem on Hue 4.4 & Solr 7.4 When you try to open a saved dashboard, the collection name is not included in the Solr call. We checked the database, collection name is saved in dashboard record on database. Might be an UI bug?

burakongun · ‎04-22-2019

I am having the same problem. I want to create a local user for minifi on a Windows server on a domain, so I check the Local User option but it still tries to use Get-ADuser module. HDF 3.4 Windows Server 2012 R2 Have you been able to solve it?

burakongun · ‎04-12-2019

I have a flow attribute which has multiple separators formetted as: header1|header2|header3~key1=value1~key2=value2~key3=value3 How can I convert this to JSON as: "headervalue1" : "header1", "headervalue2" : "header2", "headervalue3" : "header3", "key1" : "value1", "key2" : "value2", "key3" : "value3" Thanks.

burakongun · ‎11-20-2017

SOLVED: The problem was the CM's mysql database. Navigator index data filled up the disk partition that mysql data is in. Once I cleaned it CDH services worked.

burakongun · ‎11-02-2017

I couldn't figure out what went wrong with our Cloudera cluster, there are multiple errors but most important one is that I can't access Cloudera Manager web UI at https://<cmserverip>:7183/cmf/home. It throws ERR_CONNECTION_CLOSED on Chrome. If I run netstat -anp | grep 7183, it's 5000+ lines of CLOSED_WAITs with a bunch of ESTABLISHEDs all from the localhost IP * Selinux is disabled. Iptables are empty. * hostname -f returns correct name. * /etc/hosts file are all same and correct on all nodes * I also can't access any of the web UIs of the Hadoop services (such as Hue) on the cluster. They throw ERR_CONNECTION_TIMED_OUT on Chrome. * All CDH services are up and accessible through command-line. * None of logs I did find anything unusual. * I CAN access non-CDH services on the nodes like RStudio or Jupyter. I'm suspecting something failed with certification stuff or Kerberos but I don't know how to check those things. Would appreciate any help.

burakongun · ‎07-19-2017

I managed to start up the tablet servers. I checked the role log and there was an error saying block memory size is not enough so I changed it from Cloudera Manager. If anyone still has this problem, checking the role log might help.

burakongun · ‎07-12-2017

My problem is indeed that there is no log4j.properties in the process folder but it wasn't because of an unzipping process as I see. Just this in stderr.log: sed: can't read /var/run/cloudera-scm-agent/process/xxxx-accumulo16-ACCUMULO16_TSERVER/log4j.properties: No such file or directory Then log4j warnings at the bottom. I also tried removing all and setting up again like you did but faced the same problem. Tablet servers don't start up. Adding "/etc/accumulo/conf" to accumulo classpath also doesn't help. I believe I'm having a different issue but can't see it because of the logging configuration. I tried adding these properties that I found at somewhere else to Accumulo configuration on Cloudera Manager but it doesn't seem to work Tablet Server Logging Advanced Configuration Snippet (Safety Valve): log4j.rootLogger=INFO, logAppender, console log4j.appender.logAppender=org.apache.log4j.RollingFileAppender log4j.appender.logAppender.File=accumulo.log log4j.appender.logAppender.MaxFileSize=8192KB log4j.appender.logAppender.layout=org.apache.log4j.PatternLayout log4j.appender.logAppender.layout.ConversionPattern=%d [%t] %-5p %c - %m%n log4j.appender.console=org.apache.log4j.ConsoleAppender log4j.appender.console.Target=System.err log4j.appender.console.Threshold=WARN log4j.appender.console.layout=org.apache.log4j.PatternLayout log4j.appender.stdout.layout.ConversionPattern=%d{yyyy-MM-dd HH:mm:ss} %-5p %c{1}:%L - %m%n I just couldn't figure out how I can include log4j.properties in the process folder when starting the Tablet Server role.

burakongun · ‎07-11-2017

Hey, I am having the exact same problem (except my CDH is 5.8). Have you been able to solve it?

burakongun · ‎07-03-2017

I don't know if this is late but I had the similar problem (with --boundary-query) on Oozie Hue interface. I was giving the query parameter without the any quotation marks and it was working fine. It failed after I rescheduled the workflow and the reason was it was parsing only the 1st word (select), so I put the single quotes and got the error you posted. Then I removed qutoes again and it worked.

burakongun · ‎04-07-2017

@bgooley Thanks for the tip. I checked Reports Manager and Navigator Metadata Server logs and got some insights about the problem. Appearently there's an issue with Kerberos authentication. I'll post Cloudera Manager forum if I can't solve it.

burakongun · ‎03-29-2017

After working around my previous problem here, I found out that the File Browser on Cloudera Manager is not up to date, all the files & folders are old, which makes reports to not show anything useful. On the top right corner on HDFS -> File Browser -> Directory Usage page it says "The file system image was last indexed on January 26, 2017 1:55 PM". This is preventing me from generating disk usage reports and also analyzing statistics on Cloudera Navigator. I' ve found this very informing article, but couldn't really helped me. If I'm not wrong there is something wrong with namenodes are not checkpointing or not using the image they extract. Can anyone give more information about this? As it shows in the outputs I posted below, it does checkpoint every 1 hour as in the config but how can I make Cloudera Manager, Navigator, File Browser etc. to see it? I have 1 active 1 stand-by namenodes. Checkpoint directory setting: Though, I have no "/ds1/dfs/snn" folder on stand-by namenode machine. When I search for fsimage* on active namenode: 2015-08-19+17:54: 7878 ./ds1/dfs/snn/current/fsimage_0000000000000000279 2015-08-19+17:54 62 ./ds1/dfs/snn/current/fsimage_0000000000000000279.md5 2015-08-19+17:54 10272 ./ds1/dfs/snn/current/fsimage_0000000000000000510 2015-08-19+17:54 62 ./ds1/dfs/snn/current/fsimage_0000000000000000510.md5 2017-03-29+13:57 281926964 ./ds1/dfs/nn/current/fsimage_0000000000287939457 (today) 2017-03-29+13:57 281926964 ./ds2/dfs/nn/current/fsimage_0000000000287939457 2017-03-29+13:58 62 ./ds1/dfs/nn/current/fsimage_0000000000287939457.md5 2017-03-29+13:58 62 ./ds2/dfs/nn/current/fsimage_0000000000287939457.md5 2017-03-29+14:58 282426827 ./ds1/dfs/nn/current/fsimage_0000000000287980225 2017-03-29+14:58 282426827 ./ds2/dfs/nn/current/fsimage_0000000000287980225 2017-03-29+14:58 62 ./ds1/dfs/nn/current/fsimage_0000000000287980225.md5 2017-03-29+14:58 62 ./ds2/dfs/nn/current/fsimage_0000000000287980225.md5 on stand-by namenode: 2017-03-29+13:57 281926964 ./ds1/dfs/nn/current/fsimage_0000000000287939457 2017-03-29+13:57 62 ./ds1/dfs/nn/current/fsimage_0000000000287939457.md5 2017-03-29+13:57 281926964 ./ds2/dfs/nn/current/fsimage_0000000000287939457 2017-03-29+13:57 62 ./ds2/dfs/nn/current/fsimage_0000000000287939457.md5 2017-03-29+14:58 282426827 ./ds1/dfs/nn/current/fsimage_0000000000287980225 2017-03-29+14:58 282426827 ./ds2/dfs/nn/current/fsimage_0000000000287980225 2017-03-29+14:58 62 ./ds1/dfs/nn/current/fsimage_0000000000287980225.md5 2017-03-29+14:58 62 ./ds2/dfs/nn/current/fsimage_0000000000287980225.md5

burakongun · ‎03-24-2017

An update: I was mistaken on some values. The size values on HDFS file browser and returning from hdfs dfsadmin -report are supposed values. But Cloudera metrics & charts countinue to give increasing values. du -sch output on dfs folders in Linux terminal also gives big numbers. And I noticed the increase have started a couple of days before the upgrade I mentioned, so it's not likely something went wrong with the upgrade. Recently we have been informed by another HDFS user that they have been splitting the large files into smaller ones for computing performance increase(??) which had me thinking if they're splitting the combined size of TBs of data into smaller ones mostly even smaller than Block Size (128MB) and causing usage on the file system grow more than 3x. Am I correct on this estimation?

burakongun · ‎03-24-2017

@csguna I haven't tried that, and probably would not be able to.

burakongun · ‎03-24-2017

I haven't tried that, and probably would not be able to.

burakongun · ‎03-20-2017

I've done an upgrade to Cloudera Manager from 5.5.3 to 5.10.0 then upgraded CDH from 5.5.1 to 5.8.4. After these operations, I saw the disk usages of all DataNodes on Hosts->All Hosts page increased. On HDFS file browser and with CLI commands I see almost every directory has double the size before, but I noticed no difference among the file counts, types, names etc.. Same thing when I also check disk usage on Linux terminal. I am a little bit confused and need help to figure out what happened.

burakongun · ‎02-03-2017

I'm guessing I somehow managed to do the same typo twice while changing the password. Tried the combinations but no luck. CDH version is 5.5.1 There is no /var/lib/cloudera-scm-server-db folder. Here's my /etc/cloudera-scm-server/db.properties file: com.cloudera.cmf.db.type=mysql com.cloudera.cmf.db.host=hds01:3306 com.cloudera.cmf.db.name=scmdb com.cloudera.cmf.db.user=scm com.cloudera.cmf.db.password=scm_passwd The host I'm on is hdsm. I browsed for a solution and tried these: $ psql -U scm -d scmdb Password for user scm: scm_passwd psql: FATAL: password authentication failed for user "scm" $ psql -h hds01 -U scm -d scmdb psql: could not connec to server: Connection Refused Is the server running on host "hds01" and accepting TCP/IP connections on port 5432? There's mysqld up on hds01. On hds01; $ mysql -u scm -p scm_passwd -h localhost ERROR 1045 (28000): Access denied for user 'scm'@'localhost' (using password: YES) Every other username/password combination fails and I can't seem to connect to the DB whatsoever. I'm not able to reach anyone who knows the credentials, in fact I doubt there is anyone. What's the consequences if I stop mysql, start with skip-grant, edit user passwords, flush privileges then restart? Appreciate any help EDIT (SOLVED): I'm now connected to database (found out scm user is only accepted from hdsm, so installed mysql client on there) and thanks to rufusayeni's comment at this post, I solved my issue. Though, I would wanna know what if I couldn't.

Online	Offline
Last Visited	‎06-12-2019 04:50 AM

Date Registered	‎02-03-2017 01:40 AM
Last Visited	‎06-12-2019 04:50 AM
Total Messages Posted	16
Total Kudos Received	1

Re: Connection Closed when accessing the Cloudera ...

Re: [HUE][SOLR] Hue Dashboard interface problem

Re: I can not install minifi-3.1.1.0-35.msi on wi...

[NIFI] Parsing multi delimiter attribute into JSON

Re: Connection Closed when accessing the Cloudera ...

Connection Closed when accessing the Cloudera Mana...

Re: Trouble Installing Accumulo 1.6 on CDH 5.6

Re: Trouble Installing Accumulo 1.6 on CDH 5.6

Re: Trouble Installing Accumulo 1.6 on CDH 5.6

Re: Sqoop fails with "Invalid SQL type: sqlKind = ...

Re: File Browser on Cloudera Manager is not up to ...

File Browser on Cloudera Manager is not up to date

Re: HDFS disk usage doubled after upgrade

Re: HDFS disk usage doubled after upgrade

Re: HDFS disk usage doubled after upgrade

HDFS disk usage doubled after upgrade

Resetting Cloudera Manager admin password