Bug 1125301
Summary: | HA: Add heat-engine with heat-api-cloudwatch to colocation constraint | ||
---|---|---|---|
Product: | Red Hat OpenStack | Reporter: | Leonid Natapov <lnatapov> |
Component: | openstack-foreman-installer | Assignee: | Brad P. Crochet <brad> |
Status: | CLOSED ERRATA | QA Contact: | Leonid Natapov <lnatapov> |
Severity: | unspecified | Docs Contact: | |
Priority: | high | ||
Version: | 5.0 (RHEL 7) | CC: | brad, fdinitto, jguiditt, mburns, morazi, rhos-maint, yeylon |
Target Milestone: | ga | Keywords: | ZStream |
Target Release: | Installer | ||
Hardware: | Unspecified | ||
OS: | Unspecified | ||
Whiteboard: | |||
Fixed In Version: | openstack-foreman-installer-2.0.20-1.el6ost | Doc Type: | Bug Fix |
Doc Text: | Story Points: | --- | |
Clone Of: | Environment: | ||
Last Closed: | 2014-08-21 18:07:29 UTC | Type: | Bug |
Regression: | --- | Mount Type: | --- |
Documentation: | --- | CRM: | |
Verified Versions: | Category: | --- | |
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
Cloudforms Team: | --- | Target Upstream Version: | |
Embargoed: |
Description
Leonid Natapov
2014-07-31 14:05:02 UTC
So, looking at the how to, heat engine is set to active-passive unless ceilometer is configured, which it is not, yet, so I am not sure this will work as intended with our existing codebase. I think with A/P for heat-engine, this would effectively make api-cloudwatch also A/P, even though we have it A/A right now. From the howto[1] # heat-engine can do A/A but requires OS::Ceilometer::Alarm in templates # that means: # 1) ceilometer must be working before heat # 2) if somebody overrides a template can it go kaboom? # 3) let´s start basic with A/P, we can easily tune it later on. pcs resource create heat-engine systemd:openstack-heat-engine op monitor start-delay=10s The constraints are then listed as: pcs constraint order start heat-api-cloudwatch-clone then heat-engine-clone pcs constraint colocation add heat-engine with heat-api-cloudwatch Clearly the second is what you are asking for, but notice the first, which refers to heat-engine-clone, which is not what we created in the step before. If Fabio and company say it is ok to mix these, I am happy to add it, I am just concerned that the A/A setup and thus the cool constraint are required, in which case we need to have this depend on the HA ceilometer feature, which has not yet been started. [1] https://github.com/fabbione/rhos-ha-deploy/blob/master/rhos5-rhel7/mrgcloud-setup/RHOS-RHEL-HA-how-to-mrgcloud-rhos5-on-rhel7-heat-latest.txt (In reply to Jason Guiditta from comment #1) > So, looking at the how to, heat engine is set to active-passive unless > ceilometer is configured, which it is not, yet, so I am not sure this will > work as intended with our existing codebase. I think with A/P for > heat-engine, this would effectively make api-cloudwatch also A/P, even > though we have it A/A right now. From the howto[1] > > # heat-engine can do A/A but requires OS::Ceilometer::Alarm in templates > # that means: > # 1) ceilometer must be working before heat > # 2) if somebody overrides a template can it go kaboom? > # 3) let´s start basic with A/P, we can easily tune it later on. > > pcs resource create heat-engine systemd:openstack-heat-engine op monitor > start-delay=10s > > The constraints are then listed as: > > pcs constraint order start heat-api-cloudwatch-clone then heat-engine-clone > pcs constraint colocation add heat-engine with heat-api-cloudwatch > > Clearly the second is what you are asking for, but notice the first, which > refers to heat-engine-clone, which is not what we created in the step > before. If Fabio and company say it is ok to mix these, I am happy to add > it, I am just concerned that the A/A setup and thus the cool constraint are > required, in which case we need to have this depend on the HA ceilometer > feature, which has not yet been started. > > [1] > https://github.com/fabbione/rhos-ha-deploy/blob/master/rhos5-rhel7/mrgcloud- > setup/RHOS-RHEL-HA-how-to-mrgcloud-rhos5-on-rhel7-heat-latest.txt oh you are mixing up things too much. heat-engine is A/P: check. pcs constraint colocation add heat-engine with heat-api-cloudwatch does NOT make heat-api-cloudwatch A/P. it means that heat-engine should be running together with *ANY* of the instances of api-cloudwatch. Doesn't matter which one. using heat-api-cloudwatch or heat-api-cloudwatch-clone is irrelevant for the constraint location. It makes a slightly small difference in the start order by shaving a second or two. PR submitted https://github.com/redhat-openstack/astapor/pull/351 merged by cwolfe openstack-foreman-installer-2.0.20-1.el6ost [root@mac047d7b627d5a haproxy]# pcs constraint | grep heat start openstack-heat-api-clone then start openstack-heat-api-cfn-clone start openstack-heat-api-cfn-clone then start openstack-heat-api-cloudwatch-clone start openstack-heat-api-cloudwatch-clone then start openstack-heat-engine openstack-heat-api-cfn-clone with openstack-heat-api-clone openstack-heat-api-cloudwatch-clone with openstack-heat-api-cfn-clone openstack-heat-engine with openstack-heat-api-cloudwatch-clone Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. http://rhn.redhat.com/errata/RHBA-2014-1090.html |