Bug 789140

Summary:	RFE: Implement batch processing that stops on first error
Product:	Red Hat Enterprise Linux 7	Reporter:	Jaroslav Kortus <jkortus>
Component:	pcs	Assignee:	Chris Feist <cfeist>
Status:	CLOSED NOTABUG	QA Contact:	cluster-qe <cluster-qe>
Severity:	medium	Docs Contact:
Priority:	medium
Version:	7.0	CC:	abeekhof, cluster-maint, tojeline
Target Milestone:	rc	Keywords:	FutureFeature
Target Release:	7.0
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:		Doc Type:	Enhancement
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2015-04-13 05:25:15 UTC	Type:	---
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Bug Depends On:
Bug Blocks:	1205796

Description Jaroslav Kortus 2012-02-09 22:22:52 UTC

Description of problem:
Currently crm supports multiple ways how to process a batch of commands. One being feeding from stdin and another one -f parameter.

Unfortunately none of these stops when the batch contains invalid or unsuccessful command. It just keeps processing the batch until EOF.

Sometimes this is not the desired behaviour and stopping on first error would make more sense. I'd personally prefer this to be automatic for -f. Adding another option for either stopping or not stopping is also fine.

Version-Release number of selected component (if applicable):
pacemaker-1.1.6-3.el6.x86_64

How reproducible:
always

Steps to Reproduce:
$ cat commands.txt 
cib use nonexistent_CIB
configure primitive ClusterIP ocf:heartbeat:IPaddr2 params ip=192.168.100.11 cidr_netmask=32 op monitor interval=30s	
nonexistent_command

crm -f commands.txt
  
Actual results:
$ crm -f commands.txt 
ERROR: 1: nonexistent_CIB: no such shadow CIB
ERROR: 3: syntax: nonexistent_command


Expected results:
just ERROR 1  printed, exit != 0 (that's correct already).
command on 2nd line and on not executed.

Additional info:

Comment 2 Andrew Beekhof 2012-02-10 13:18:01 UTC

Chris, something to keep in mind when you're implementing pcs.

Comment 3 Andrew Beekhof 2012-08-16 03:51:23 UTC

Moving to the pcs component

Comment 7 Tomas Jelinek 2015-03-12 14:23:22 UTC

In pcs the -f flag works differently than in crm.  It does not specify a file containing a batch of commands, it points to a file containing a CIB pcs should operate on.  If an operation fails, pcs returns a non-zero exit status indicating a failure occurred.  This can be used to detect failures in a script calling pcs.

I spoke with Jaroslav and he agrees this is not an issue in pcs.

Comment 8 Chris Feist 2015-04-13 05:25:15 UTC

Closing due to comment #7.  Users can work on an offline cib, if they run into errors than they can decide to not load back into the system.