[jira] [Commented] (CASSANDRA-7950) Output of nodetool compactionstats and compactionhistory does not work well with long keyspace and column family names.
[
https://issues.apache.org/jira/browse/CASSANDRA-7950?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15124553#comment-15124553
]
Yuki Morishita commented on CASSANDRA-7950:
---
I created patch for trunk that introduces {{TableBuilder}} (similar to what we
do now in {{compactionstats}}) and uses it for {{compactionhistory}},
{{compactionstats}}, and {{listsnapshots}}.
||branch||testall||dtest||
|[7950|https://github.com/yukim/cassandra/tree/7950]|[testall|http://cassci.datastax.com/view/Dev/view/yukim/job/yukim-7950-testall/lastCompletedBuild/testReport/]|[dtest|http://cassci.datastax.com/view/Dev/view/yukim/job/yukim-7950-dtest/lastCompletedBuild/testReport/]|
> Output of nodetool compactionstats and compactionhistory does not work well
> with long keyspace and column family names.
> -
>
> Key: CASSANDRA-7950
> URL: https://issues.apache.org/jira/browse/CASSANDRA-7950
> Project: Cassandra
> Issue Type: Improvement
> Components: Tools
> Environment: CentOS 5, 64bit, Oracle JDK 7, DSE
>Reporter: Eugene
>Assignee: Michael Shuler
>Priority: Minor
> Labels: lhf
> Fix For: 2.1.x
>
> Attachments: 7950.patch, nodetool-examples.txt
>
>
> When running these commands:
> nodetool compactionstats
> nodetool compactionhistory
> The output can be difficult to grok due to long keyspace names, column family
> names, and long values. I have attached an example.
> It's difficult for both humans and grep/sed/awk/perl to read.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (CASSANDRA-7950) Output of nodetool compactionstats and compactionhistory does not work well with long keyspace and column family names.
[
https://issues.apache.org/jira/browse/CASSANDRA-7950?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14730579#comment-14730579
]
Eugene commented on CASSANDRA-7950:
---
Thank you to everyone that has worked on this so far.
Will this, by chance, also fix the output of other commands, for example
`nodetool listsnapshots`?
I created a snapshot with a long tag
(some-really-long-snapshot-name-geezey-petes). Ended up with the following:
{code}
# nodetool listsnapshots
Snapshot Details:
Snapshot name Keyspace nameColumn family name
True size Size on disk
some-really-long-snapshot-name-geezey-petesgalactaron
replicatorstide 0 bytes570.68 MB
some-really-long-snapshot-name-geezey-petesdse_system leases
0 bytes13 bytes
some-really-long-snapshot-name-geezey-petesdse_system
real_leaders 0 bytes13 bytes
some-really-long-snapshot-name-geezey-petesdse_system
encrypted_keys 0 bytes13 bytes
some-really-long-snapshot-name-geezey-petesdse_system
registered_leaders 0 bytes13 bytes
{code}
> Output of nodetool compactionstats and compactionhistory does not work well
> with long keyspace and column family names.
> -
>
> Key: CASSANDRA-7950
> URL: https://issues.apache.org/jira/browse/CASSANDRA-7950
> Project: Cassandra
> Issue Type: Bug
> Environment: CentOS 5, 64bit, Oracle JDK 7, DSE
>Reporter: Eugene
>Assignee: Michael Shuler
>Priority: Minor
> Labels: lhf
> Fix For: 2.1.x
>
> Attachments: 7950.patch, nodetool-examples.txt
>
>
> When running these commands:
> nodetool compactionstats
> nodetool compactionhistory
> The output can be difficult to grok due to long keyspace names, column family
> names, and long values. I have attached an example.
> It's difficult for both humans and grep/sed/awk/perl to read.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (CASSANDRA-7950) Output of nodetool compactionstats and compactionhistory does not work well with long keyspace and column family names.
[ https://issues.apache.org/jira/browse/CASSANDRA-7950?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14288820#comment-14288820 ] Anuja Mandlecha commented on CASSANDRA-7950: @Brandon: We will have to definitely add a delimeter option (-- delimeter )to the nodetool command wherein the output will be displayed seperated by that delimiter and on which cut,grep etc commands can work. @Eugene: That looks like a good option. Will need to investigate more on this. Output of nodetool compactionstats and compactionhistory does not work well with long keyspace and column family names. - Key: CASSANDRA-7950 URL: https://issues.apache.org/jira/browse/CASSANDRA-7950 Project: Cassandra Issue Type: Bug Environment: CentOS 5, 64bit, Oracle JDK 7, DSE Reporter: Eugene Priority: Minor Labels: lhf Fix For: 2.0.13 Attachments: nodetool-examples.txt When running these commands: nodetool compactionstats nodetool compactionhistory The output can be difficult to grok due to long keyspace names, column family names, and long values. I have attached an example. It's difficult for both humans and grep/sed/awk/perl to read. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (CASSANDRA-7950) Output of nodetool compactionstats and compactionhistory does not work well with long keyspace and column family names.
[ https://issues.apache.org/jira/browse/CASSANDRA-7950?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14283576#comment-14283576 ] Anuja Mandlecha commented on CASSANDRA-7950: I am taking up this bug. Output of nodetool compactionstats and compactionhistory does not work well with long keyspace and column family names. - Key: CASSANDRA-7950 URL: https://issues.apache.org/jira/browse/CASSANDRA-7950 Project: Cassandra Issue Type: Bug Environment: CentOS 5, 64bit, Oracle JDK 7, DSE Reporter: Eugene Priority: Minor Labels: lhf Fix For: 2.0.12 Attachments: nodetool-examples.txt When running these commands: nodetool compactionstats nodetool compactionhistory The output can be difficult to grok due to long keyspace names, column family names, and long values. I have attached an example. It's difficult for both humans and grep/sed/awk/perl to read. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (CASSANDRA-7950) Output of nodetool compactionstats and compactionhistory does not work well with long keyspace and column family names.
[ https://issues.apache.org/jira/browse/CASSANDRA-7950?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14283760#comment-14283760 ] Anuja Mandlecha commented on CASSANDRA-7950: Hello, There are few points to be considered here: 1. The max char limit for a keyspace and cf name in CQL is 48 chars each. 2. the format specifiers mentioned in nodetool compactionhistory output in code is having 19 chars for keyspace and 29 chars for cfname (%-19s%-29s). Similar is the case with compactionstats command. Hence if the keyspace or cfnames exceed above these limits, there will not be any space between it's value and the next column value. To resolve this issue, we can have two approaches: 1. Keep existing format specifiers and add a tab after each column. 2. Remove existing format specifiers and display complete column values with only seperated by tab. I would suggest approach 1 so as to maintain human readability and make the output parsable. Output of nodetool compactionstats and compactionhistory does not work well with long keyspace and column family names. - Key: CASSANDRA-7950 URL: https://issues.apache.org/jira/browse/CASSANDRA-7950 Project: Cassandra Issue Type: Bug Environment: CentOS 5, 64bit, Oracle JDK 7, DSE Reporter: Eugene Priority: Minor Labels: lhf Fix For: 2.0.12 Attachments: nodetool-examples.txt When running these commands: nodetool compactionstats nodetool compactionhistory The output can be difficult to grok due to long keyspace names, column family names, and long values. I have attached an example. It's difficult for both humans and grep/sed/awk/perl to read. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (CASSANDRA-7950) Output of nodetool compactionstats and compactionhistory does not work well with long keyspace and column family names.
[ https://issues.apache.org/jira/browse/CASSANDRA-7950?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14284190#comment-14284190 ] Jonathan Ellis commented on CASSANDRA-7950: --- Doesn't using tabs mean different-length names will not line up if they differ by more than 8 characters? Output of nodetool compactionstats and compactionhistory does not work well with long keyspace and column family names. - Key: CASSANDRA-7950 URL: https://issues.apache.org/jira/browse/CASSANDRA-7950 Project: Cassandra Issue Type: Bug Environment: CentOS 5, 64bit, Oracle JDK 7, DSE Reporter: Eugene Priority: Minor Labels: lhf Fix For: 2.0.13 Attachments: nodetool-examples.txt When running these commands: nodetool compactionstats nodetool compactionhistory The output can be difficult to grok due to long keyspace names, column family names, and long values. I have attached an example. It's difficult for both humans and grep/sed/awk/perl to read. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (CASSANDRA-7950) Output of nodetool compactionstats and compactionhistory does not work well with long keyspace and column family names.
[ https://issues.apache.org/jira/browse/CASSANDRA-7950?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14284206#comment-14284206 ] Eugene commented on CASSANDRA-7950: --- Might be worth looking into how the column command does it. e.g.: cat /proc/meminfo cat /proc/meminfo | column -t At the very least if these nodetool commands had proper delimitation (a single space between fields) their output could be easily parsed by utilities and the column command could be used manually to correct the output. Output of nodetool compactionstats and compactionhistory does not work well with long keyspace and column family names. - Key: CASSANDRA-7950 URL: https://issues.apache.org/jira/browse/CASSANDRA-7950 Project: Cassandra Issue Type: Bug Environment: CentOS 5, 64bit, Oracle JDK 7, DSE Reporter: Eugene Priority: Minor Labels: lhf Fix For: 2.0.13 Attachments: nodetool-examples.txt When running these commands: nodetool compactionstats nodetool compactionhistory The output can be difficult to grok due to long keyspace names, column family names, and long values. I have attached an example. It's difficult for both humans and grep/sed/awk/perl to read. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (CASSANDRA-7950) Output of nodetool compactionstats and compactionhistory does not work well with long keyspace and column family names.
[ https://issues.apache.org/jira/browse/CASSANDRA-7950?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14284229#comment-14284229 ] Brandon Williams commented on CASSANDRA-7950: - If the problem is reduced to simply being able to parse the output, a better option would be to add a csv output mode. Output of nodetool compactionstats and compactionhistory does not work well with long keyspace and column family names. - Key: CASSANDRA-7950 URL: https://issues.apache.org/jira/browse/CASSANDRA-7950 Project: Cassandra Issue Type: Bug Environment: CentOS 5, 64bit, Oracle JDK 7, DSE Reporter: Eugene Priority: Minor Labels: lhf Fix For: 2.0.13 Attachments: nodetool-examples.txt When running these commands: nodetool compactionstats nodetool compactionhistory The output can be difficult to grok due to long keyspace names, column family names, and long values. I have attached an example. It's difficult for both humans and grep/sed/awk/perl to read. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
