Bug 1330774
| Summary: | RFE: Add a --simulate switch to pcs | ||
|---|---|---|---|
| Product: | Red Hat Enterprise Linux 8 | Reporter: | Michele Baldessari <michele> |
| Component: | pcs | Assignee: | Tomas Jelinek <tojeline> |
| Status: | CLOSED WONTFIX | QA Contact: | cluster-qe <cluster-qe> |
| Severity: | medium | Docs Contact: | |
| Priority: | medium | ||
| Version: | 8.2 | CC: | cfeist, cluster-maint, fdinitto, idevat, kgaillot, omular, tojeline |
| Target Milestone: | rc | Keywords: | FutureFeature |
| Target Release: | 8.2 | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | Enhancement | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2020-11-01 03:02:58 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: | |||
This is not easy to do because many commands push the CIB several times and therefore there is not a single point at the end of the commands to put running crm_simulate to. These commands need to be overhauled to only push the CIB once at their end so pcs can ask pacemaker to simulate it. We can add a generic simulate mechanism to the new architecture and connect it to commands being moved to the new architecture. After evaluating this issue, there are no plans to address it further or fix it in an upcoming release. Therefore, it is being closed. If plans change such that this issue will be fixed in an upcoming release, then the bug can be reopened. FYI: crm_simulate in the Pacemaker upstream master branch (expected to land in upstream 2.1.0 and RHEL 8.5/9.0beta) now has an --output-as=xml option, to obtain reliably parseable output (the text output may change from release to release). |
The idea behind this request is to add a --simulate switch to pcs that will print out the actions that the cluster will take without actually doing it. The background for this is that for an operator it is quite hard to predict what the cluster will do after a "pcs ..." action. Especially in complex clusters (think of the control plane in openstack) with lots of constraints and resources this becomes an especially daunting task. An example output could be: $ pcs --simulate resource disable foo The following actions will be undertaken: 1. resource A (depends on B) will stop on all nodes 2. resource B (depends on foo) will stop on all nodes One of the side effects here would be to reuse crm_simulate and make it completely user-friendly.