Creating a new mixed group from the bundle deployment wizzard
by Jiri Kremser
Hi,
this issue is based on bug 845363. The problem is that if user creates a mixed group from the wizzard when creating a new bundle deployment. The new group is not populated in the available groups. It is obviously caused by the assumption on bundle deployments that only compatible groups with a resource type supporting the bundle deployment.
I would propose the solution: if user tries to create a new group from that wizzard, it doesn't allow it if the conditions are not met (mixed or res. type don't support depl.), and user is notified why it is so.
In the code it would mean changes in GetDestinationStep.QuickGroupCreateWizard (1 new method) and 1 new method on BundleGWTService and BundleManager (taking the set of resource ids and checking the conditions).
JK
11 years, 3 months
Bug 887320: OutOfMemoryError when adding new Deployment child resource on AS7
by Thomas Segismont
Hi all,
I'm writing to you about this bug:
https://bugzilla.redhat.com/show_bug.cgi?id=887320
For those not aware of it, an OutOfMemoryError is raised on the agent
when you try to add a child resource of type deployment to a managed AS7
instance.
It happens because the AS7 plugin uses HttpUrlConnection to communicate
with AS7 HTTP management interface.
Basically, HttpUrlConnection waits for the request payload to be fully
loaded in memory before it tries to contact the server. It does that to
be able to resend the request payload if the server ask for redirection
or start an authentication challenge.
While this is acceptable for small payloads, it's not suitable for large
file uploads (in our case a 42M WAR upload triggers the OutOfMemoryError).
For large file uploads, HttpUrlConnection provides a streaming mode. But
in streaming mode it does neither support redirection nor
authentication. I tried to use streaming mode to make a fix yesterday
and it's really not working (BTW, sorry for the patch revert waltz).
I can see two options:
#1 Keep HttpUrlConnection as communication layer
In this case, the streaming mode is mandatory. This means we should
maintain our own authentication challenge code and set authentication
headers manually.
#2 Use HttpClient as communication layer
Personally, I prefer #2
Importing HttpClient in AS7 plugin will increase the plugin
dependencies list. But I think it's a small price to pay compared to the
efforts we would spend maintaining our own authentication and
redirection code.
What do you think?
Thanks
Thomas
11 years, 3 months
Metrics Migration Tool - Cassandra
by Stefan Negrea
Hello Everybody,
With new Cassandra based metrics storage system, existing data will need to be migrated out of existing SQL storage to Cassandra. I put together a simple design document for the data migration tool that will be delivered with RHQ when the new metrics storage system is completed.
My initial thoughts about the tool are:
1) Restartable migration process, this will be especially useful for users with large amounts of metrics.
2) Favour robustness over performance because this migration will be done only once.
3) Make the tool external to the product (not part of the installation), so user can do the migration ahead of the upgrade.
Here is the full design document:
https://docs.jboss.org/author/display/RHQ/Metrics+Data+Migration+-+Design
Feedback is more than welcomed since this tool is still in the early planning stages.
Thank you,
Stefan Negrea
Software Engineer
11 years, 3 months
minor improvement to welcome message portlet
by Jay Shaughnessy
I never liked the welcome portlet and it's useless canned message about
RHQ. It's likely you know what RHQ is about if you're running it
already. I pushed a change that instead cleans it up and offers a few
basic links. I think even in the current form it's more useful than
before, or minimally less useless.
The general idea is to give people some links for things that may be
useful to do. The UI gets beat up a lot because it's so general, not
task-focused. This isn't really that but it hints at it by giving a
first time user an option to "Import newly discovered resources" as a
link to the ADQ.
Anyway, something to think about. Also, this is no longer localized, so
if we like it, please localize.
-Jay
11 years, 3 months
RHQ 4.6 beta
by Heiko W.Rupp
Hello,
we have just uploaded a snapshot from our internal build system (*) as a beta release of RHQ 4.6 to
source forge.
This build contains all the changes we made to swap out the old as4 as the underlying app server and
to replace it with as7. There are other changes like enhancements to the rest-api, Korean translations (**) ,
GWT updates or the new new installer.
Please download the release, try it and give us feedback. We intend to release a final 4.6 release
around the end of the month, so your feedback is very valuable to us.
You can get the release from https://sourceforge.net/projects/rhq/files/rhq/rhq-4.6.0-beta/
Follow the instructions on that wiki to run the installer: https://docs.jboss.org/author/display/RHQ/Installing+RHQ+on+JBossAS+7
Heiko on behalf of the RHQ team
*) We use the same system to build the releases, but just did not go through the steps to tag
the release and create a "named" build.
**) Those were unfortunately missing from the first version of the download, that is now fixed.
--
Reg. Adresse: Red Hat GmbH, Technopark II, Haus C,
Werner-von-Siemens-Ring 14, D-85630 Grasbrunn
Handelsregister: Amtsgericht München HRB 153243
Geschaeftsführer: Mark Hegarty, Charlie Peters, Michael Cunningham, Charles Cachera
11 years, 3 months
Discrepancy between REST and remote API
by Lukas Krejci
Hi all,
REST has been gaining functionality that is not available through our remote
API. Lately, it is the ability to create alert definitions, but it also can
create dynagroups, afair, which is something remote API can't do either.
While I welcome these additions and enabling remote users to achieve such
things, I am also very concerned about fragmenting the capabilities of the two
remote endpoints of RHQ.
I really do think that all the endpoints should be equal in their
capabilities. Not necessarily the APIs or approaches to getting the data, but
the users should be able to achieve the same things regardless of the endpoint
they use.
For this I think that ideally REST should exclusively use the remote EJBs to
achieve its thing but that's not the case today. Because we are using local
EJBs in REST, we are free to diverge as much as we want from the remote API,
which I think is ultimately wrong and will only cause us problems when REST
becomes stable and users will start asking questions why is this or that
possible in REST and not in remote API (or vice versa).
Just my 2c, but I haven't heard any solid opinion on this problem or how to
resolve it which is why I'm bringing it up here.
Lukas
11 years, 3 months
Metrics Collection - Help
by Stefan Negrea
Hello Everybody,
I just worked on estimating the amount of numeric metrics data collected by RHQ. I would appreciate a little feedback on the work I've done. This will help me get a better idea on the ratios of agent:raw metrics and agent:aggregate metric collected on real environments.
For all RHQ users with a Postgres backend, could you please run the query below on your RHQ database and reply with the result? The result is a set of three numbers: the number of agents, total number of aggregate metrics, and total number of raw metrics.
All the feedback is greatly appreciated, even if you are developer, plugin writer, casual tester or you just installed RHQ for a simple test.
The query:
SELECT COUNT(*) FROM rhq_agent UNION ALL
SELECT (
(SELECT COUNT(*) FROM rhq_measurement_data_num_1h) +
(SELECT COUNT(*) FROM rhq_measurement_data_num_6h) +
(SELECT COUNT(*) FROM rhq_measurement_data_num_1D) ) UNION ALL
SELECT (
(SELECT COUNT(*) FROM rhq_meas_data_num_r00) +
(SELECT COUNT(*) FROM rhq_meas_data_num_r02) +
(SELECT COUNT(*) FROM rhq_meas_data_num_r03) +
(SELECT COUNT(*) FROM rhq_meas_data_num_r04) +
(SELECT COUNT(*) FROM rhq_meas_data_num_r05) +
(SELECT COUNT(*) FROM rhq_meas_data_num_r06) +
(SELECT COUNT(*) FROM rhq_meas_data_num_r07) +
(SELECT COUNT(*) FROM rhq_meas_data_num_r08) +
(SELECT COUNT(*) FROM rhq_meas_data_num_r09) +
(SELECT COUNT(*) FROM rhq_meas_data_num_r10) +
(SELECT COUNT(*) FROM rhq_meas_data_num_r11) +
(SELECT COUNT(*) FROM rhq_meas_data_num_r12) +
(SELECT COUNT(*) FROM rhq_meas_data_num_r13) +
(SELECT COUNT(*) FROM rhq_meas_data_num_r14))
Here is what I got when I ran the query on my local box:
1
7285
18617
Thank you,
Stefan Negrea
Software Engineer
11 years, 3 months
High Availability feature testing results
by Thomas Segismont
Hi all,
I have finished to test the new high availability screens (GWT).
I did not find any HA regression, only a few (minor) issues in the GUI:
*affinity group screen does not show server members (but they are
actually saved)
*affinity group agent membership pop-up does not show all available agents
I'm not sure if it's really a bug (maybe it was the previous behavior)
* partition events screen does not show NORMAL->DOWN events for RHQ servers
Again, maybe it was the previous behavior
Regarding the testing method I used Vagrant[1] and a small set of
(dirty) Puppet manifests to create the virtual machines on my computer:
* 1 node with Postgres database
* 2 nodes with RHQ server
* 4 nodes with RHQ agent
I made an RHQ build with dist and dev profiles activated and installed
the servers from the distribution zip and the agents from the
http://[host]:[port]/agentupdate/download URL (following this wiki page
https://docs.jboss.org/author/display/RHQ/Installing+RHQ+on+JBossAS+7)
Regards
Thomas
[1] http://www.vagrantup.com/
11 years, 3 months
Wrong httpclient version in agent SAR POM file?
by Thomas Segismont
Hi all,
The POM of the agent SAR project references commons-httpclient in
version 2.0.2.
<dependency>
<groupId>commons-httpclient</groupId>
<artifactId>commons-httpclient</artifactId>
<!-- make sure you match that which is used by standalone agent! -->
<version>2.0.2</version>
</dependency>
As you can see, the comments suggests the specified version should match
the version used by standalone agent but everywhere else in the project
POM files we use 3.0.1.
Am I missing something? Obviously this does not prevent the agent to run
as expected since these lines have not changed since (at least) 2009.
Thanks and regards,
Thomas
11 years, 3 months