How to get disk usage on individual Project Area?

We use DB2 as the backend database handling RTC repository storage.
Is there a RTC way (command) to figure out the disk usage of individual RTC project area?

This will help us to prepeare migration and plan for future growth.

I guess I need to raise an ehancement request to get some attension or answer :twisted:

We use DB2 as the backend database handling RTC repository storage.
Is there a RTC way (command) to figure out the disk usage of individual RTC project area?

This will help us to prepeare migration and plan for future growth.

Hi

The server reports should show you the size of the data in the repository. Go to your project via the webUI, select Reports and look for the repository reports such as Repository Metrics. That tells you the size of the repository.

anthony

Hi

The server reports should show you the size of the data in the repository. Go to your project via the webUI, select Reports and look for the repository reports such as Repository Metrics. That tells you the size of the repository.

anthony

Anthony,
Thanks for the response.
I checked the reports you pointed but they all look the same across project areas, so I think it is not for individual Project Area, but a view of the whole repository.

It does give disk usage for different type of items, like work items, scm, etc.. But again it is the overall usage across all project areas in the same repository.

What I am looking for is a way to tell the disk usage for individual Project Area.

Hi

The server reports should show you the size of the data in the repository. Go to your project via the webUI, select Reports and look for the repository reports such as Repository Metrics. That tells you the size of the repository.

anthony

Anthony,
Thanks for the response.
I checked the reports you pointed but they all look the same across project areas, so I think it is not for individual Project Area, but a view of the whole repository.

It does give disk usage for different type of items, like work items, scm, etc.. But again it is the overall usage across all project areas in the same repository.

What I am looking for is a way to tell the disk usage for individual Project Area.

My apologies - I did not read your original question carefully.

You may be able to create a new report for projects areas - perhaps one of the people in the repors team can advise us.

I would create a new work item on this anyway - it's a very useful idea.

regards

anthony

You may be able to create a new report for projects areas - perhaps one of the people in the repors team can advise us.

I would create a new work item on this anyway - it's a very useful idea.

regards

anthony

Thanks, Anthony,
I'll create a work item on this.

https://jazz.net/jazz/web/projects/Rational%20Team%20Concert#action=com.ibm.team.workitem.viewWorkItem&id=105686

My enhancemnt request was rejected with the following comments:
Unfortunately this is not possible. Items in the repository are not necessarily even associated with particular project areas. If they are, we don't know how to tell, since the items are opaque to us (we don't have an understanding of the structure of each of them in a generic way to determine which project area they "belong" to). Therefore, the data that we aggregate in the data warehouse for this sizing information (on which the report is based) can't contain project information.

But there are different opinions:

You could do this by hooking into the read permission support, which links every item to a context that is either a project area or a contributor. IItem#getContextId() gives you the UUID of the project area for most item types.

SCM data might be problematic, though, since I believe they use a custom mechanism for read permissions. Still, querying for items whose context ID equals the UUID of the project area you're interested in might get you a long way.


So should I reopen the work item to request reconsidering it?

To monitor database sizes/capacity (db2 internal) I use:


db_sizes

#!/bin/ksh
for DB in $(db2 list database directory |grep "Database name" | awk -F"=" '{print $2}' |sort ); do
db2 connect to $DB > /dev/null 2>&1
#db2 -x "select tabschema,tabname,colname,typename from syscat.columns where colname = '$what'" > /tmp/list
print -n "$DB,"
db2 "CALL GET_DBSIZE_INFO(?, ?, ?, 0)" |grep "Parameter Value" | awk -F ":" '{printf "%s,", $2}'
print

done

I neaten it up (convert to GB) by sending that to this awk script:

awk -F "," '{printf "%s,%2.2f,%2.2f\n", $1,$3/1073741824.,$4/1073741824.}'

Gives me something like this:

Database sizes
DB Name,Used(Gb),Capacity(Gb)
RTCDB,21.14,27.70
RTCDB10,5.42,5.82
RTCDB11,2.13,2.84
RTCDB12,6.81,8.25
RTCDB15,2.03,2.22
RTCDB2,2.55,3.49
RTCDB3,2.76,4.74
RTCDB4,1.73,4.24
RTCDB5,2.40,4.88
RTCDB6,3.96,5.79
RTCDB7,1.34,2.67
RTCDB8,4.96,5.12
RTCDB9,1.49,2.67

My enhancemnt request was rejected with the following comments:
Unfortunately this is not possible. Items in the repository are not necessarily even associated with particular project areas. If they are, we don't know how to tell, since the items are opaque to us (we don't have an understanding of the structure of each of them in a generic way to determine which project area they "belong" to). Therefore, the data that we aggregate in the data warehouse for this sizing information (on which the report is based) can't contain project information.

But there are different opinions:

You could do this by hooking into the read permission support, which links every item to a context that is either a project area or a contributor. IItem#getContextId() gives you the UUID of the project area for most item types.

SCM data might be problematic, though, since I believe they use a custom mechanism for read permissions. Still, querying for items whose context ID equals the UUID of the project area you're interested in might get you a long way.


So should I reopen the work item to request reconsidering it?

To monitor database sizes/capacity (db2 internal) I use:


db_sizes

#!/bin/ksh
for DB in $(db2 list database directory |grep "Database name" | awk -F"=" '{print $2}' |sort ); do
db2 connect to $DB > /dev/null 2>&1
#db2 -x "select tabschema,tabname,colname,typename from syscat.columns where colname = '$what'" > /tmp/list
print -n "$DB,"
db2 "CALL GET_DBSIZE_INFO(?, ?, ?, 0)" |grep "Parameter Value" | awk -F ":" '{printf "%s,", $2}'
print

done

I neaten it up (convert to GB) by sending that to this awk script:

awk -F "," '{printf "%s,%2.2f,%2.2f\n", $1,$3/1073741824.,$4/1073741824.}'

Gives me something like this:

Database sizes
DB Name,Used(Gb),Capacity(Gb)
RTCDB,21.14,27.70
RTCDB10,5.42,5.82
RTCDB11,2.13,2.84
RTCDB12,6.81,8.25
RTCDB15,2.03,2.22
RTCDB2,2.55,3.49
RTCDB3,2.76,4.74
RTCDB4,1.73,4.24
RTCDB5,2.40,4.88
RTCDB6,3.96,5.79
RTCDB7,1.34,2.67
RTCDB8,4.96,5.12
RTCDB9,1.49,2.67

Kevin,
Thnaks for the help.
I believe usually DB2 databse is matching to RTC repository, not project area and normally one RTC repository contains multiple project areas.
Samilar to your method, we hav ways to tell the size of the individual DB2 database's size, which RTC's metric reports also do, but we want to know how the individualproject area is using the disk.

And the good news is that Jazz team has repoened my enhancement request and plan to make it available in 3.0.

I shared for those that haven't figured this out already :-)

Kevin,
Thnaks for the help.
I believe usually DB2 databse is matching to RTC repository, not project area and normally one RTC repository contains multiple project areas.
Samilar to your method, we hav ways to tell the size of the individual DB2 database's size, which RTC's metric reports also do, but we want to know how the individualproject area is using the disk.

And the good news is that Jazz team has repoened my enhancement request and plan to make it available in 3.0.