About slm

slm · ‎06-12-2017

HBase shell also includes an alternative command to `truncate` called `truncate_preserve` which can be used to drop data within a table but still maintain the metadata about the table contained in hbase:meta. You can also use the `status 'detailed'` command from `hbase shell` to see the split/presplit regions. For example: $ /opt/bin/hbshell.bash <<< "truncate_preserve 'dev06ostgarnerh:contact'" $ /opt/bin/hbshell.bash <<< "status 'detailed'" | grep contact "dev06ostgarh:contact,600,1497277520170.a17b52eec2a9df78bbc175a5238a13f8." "dev06ostgarh:contact,800,1497277520170.275ff20d4c4036163ee90a94092b47e6." "dev06ostgarh:contact,A00,1497277520170.e94b5202f2cbb100dd74e858f772573e." "dev06ostgarh:contact,E00,1497277520170.7d9f1b140431d2a975ee35e716b9337e." "dev06ostgarh:contact,,1497277520170.ed4ad8c3fb40598bb51d2ad91dee6973." "dev06ostgarh:contact,100,1497277520170.43378ea32c0b3bf64dd697b62b87b23c." "dev06ostgarh:contact,200,1497277520170.be10a9e051b0f9a4f7c460c1b36969d7." "dev06ostgarh:contact,F00,1497277520170.64f57d24d29058f238b1addb08c603a8." "dev06ostgarh:contact,300,1497277520170.6244d798c326984e47a577f579645487." "dev06ostgarh:contact,500,1497277520170.c9b1ed0633850eb5086a32049a5597d5." "dev06ostgarh:contact,C00,1497277520170.bb446488db4bb788248f41551db878f5." "dev06ostgarh:contact,D00,1497277520170.b2d2bc1febc9206df27e05a976066027." "dev06ostgarh:contact,400,1497277520170.b61ffc5cdfda59f96b47d83f8aa46134." "dev06ostgarh:contact,700,1497277520170.e4e1cab56f430344c71e24fa6673b55f." "dev06ostgarh:contact,900,1497277520170.1cb1067053fe203426e0e9e2c00bb37f." "dev06ostgarh:contact,B00,1497277520170.5867477ee49a19f6dafd0ff60f9b8f6f."

slm · ‎07-27-2016

http://hortonworks.com/blog/odpi-core-hdp-hadoop-core/. Looks with the above that Phoenix would be part of the extended HDP services. > Extended HDP services: Extended HDP services relates to the services that access the data stored in HDP including Hive, HBase, Storm, Spark and more. Extended services provide customers with the flexibility to uptake the innovation that is delivered on top of the common core, based on their own requirements and schedule. With Extended services, customers would be able to easily uptake the new functionality without causing any disruption to their core platform.

slm · ‎07-27-2016

As we've been evaluating the use of Phoenix + HBase in HW 2.3, we came upon the conclusion that in the current bundled version of Phoenix, 4.4, that namespaces are not supported. The upstream JIRAs are confusing but it looks like Phoenix will not support namespaces + permissions until 4.7-4.8. Does this sound like a correct conclusion? If so what are my options in terms of Phoenix? upgrade it within HW 2.3 (assuming this is not an option)? work around it somehow? Move Phoenix tables manually out of default NS into our applications? How? This is what we seen when enabling Phoenix on our system. The SYSTEM.* tables are Phoenix's. Ours are in the ns: namespace. Version 1.1.2.2.3.6.0-3796, r2873b074585fce900c3f9592ae16fdd2d4d3a446, Thu Jun 23 16:29:31 UTC 2016 hbase(main):001:0> list TABLE SYSTEM.CATALOG SYSTEM.FUNCTION SYSTEM.SEQUENCE SYSTEM.STATS ambarismoketest ns:contact ns:counters ns:counters-backup ns:lists ns:logins ns:modelLogs ns:models ns:reportLogs ns:users References HBase namespaces surfaced in phoenix Support HBase non-default Namespace Phoenix Schema should mapping to HBase Namespace

slm · ‎07-23-2016

I've seen it in as far back as 2.1 that's why I was surprised it was missing in my install.

slm · ‎07-22-2016

Yeah perhaps it's a bug. I saw no indications of the Phoenix jar files when I did a find on a 2.3 system. So I did a yum install phoenix and they got installed. This box was pretty basic with just HDFS, Zookeeper, & HBase. I'll spin another env up and see if I encounter the issue again. Thanks for the help.

slm · ‎07-22-2016

This seemed a little odd to me that I had to do an actual `yum install phoenix` command to install Phoenix on HDP 2.3. In HDP 2.2 this was included by default (from what I remember) when I installed the HBase role. I suspect it was done given the size of Phoenix when I installed it being ~250MB, but this felt unusual to me given I can literally do everything else through the Ambari dashboard. Why this 2nd class treatment for Phoenix? Are there plans to install Phoenix as a actual role or will the installation be like this going forward for this particular Hadoop project?

slm · ‎07-21-2016

We recently started implementing HBase namespaces + ACLs and have run into an issue. According to the docs: http://hbase.apache.org/0.94/book/ops.snapshots.html: 14.8.7. Snapshots operations and ACLs If you are using security with the AccessController Coprocessor (See Section 8.2, “Access Control”), only a global administrator can take, clone, or restore a snapshot, and these actions do not capture the ACL rights. This means that restoring a table preserves the ACL rights of the existing table, while cloning a table creates a new table that has no ACL rights until the administrator adds them. Our application requires the ability to take a snapshot of a specific table, clone it, and then Questions Why does the snapshot mechanism require this high level access to function? Is this something that will change over time or is this the design and it's being done this way for a specific purpose?

slm · ‎07-15-2016

It appears as though adding permissions on just the namespace is not sufficient for allowing a user access to the tables within it. I had to cascade the permissions to the tables themselves like so in an hbase shell: list.each {|t| grant 'our_apps_user','RWCXA',t} I did the above using the included hbase SPN in the hbase.headless.keytab. $ kinit -kt /etc/security/keytabs/hbase.headless.keytab hbase-<servername>@<REALM> $ hbase shell ...above command...

slm · ‎07-15-2016

Sorry that was a mistake when I sanitized the output taking work specific details out. The name is our_apps_user and it's not a typo.

slm · ‎07-15-2016

Found my own answer after Googling a bit more in Community Connections here: https://community.hortonworks.com/articles/14463/auth-to-local-rules-syntax.html. [n:string] Indicates a matching rule where n declares the number of expected components in the principal. Components are separated by a /, where a user account has one component (ambari-qa) and a service account has two components (nn/fqdn). The string value declares how to reformat the value to be used in the rest of the expression. The placeholders are as follows: $0 - realm $1 - 1st component $2 - 2nd component

Online	Offline
Last Visited	‎01-31-2019 05:56 AM

Member Since	‎12-15-2015 01:43 PM
Last Visited	‎01-31-2019 05:56 AM
Posts	39
Kudos received	21

Cloudera Community

Cloudera Community

Re: Granting a Kerberos principal permissions on a...

Re: I find the auth_to_local documentation confusi...

Re: Server Error when attempting to setup Kerberos

Re: How can I stop the entire HDP stack via the CL...

Re: HBASE - truncate table/region server

Re: Can Phoenix v4.4 utilize HBase namespaces?

Can Phoenix v4.4 utilize HBase namespaces?

Re: Why can I not install Phoenix through Ambari d...

Re: Why can I not install Phoenix through Ambari d...

Why can I not install Phoenix through Ambari direc...

Why is global admin permissions necessary to creat...

Re: Granting a Kerberos principal permissions on a...

Re: Granting a Kerberos principal permissions on a...

Re: I find the auth_to_local documentation confusi...