Mailing List Archive

>>> Kristoffer Grönlund<kgronlund@suse.com> schrieb am 16.02.2015 um 09:15 in
Nachricht <87oaougtlw.fsf@krigpad.kri.gs>:
> Vladislav Bogdanov <bubble@hoster-ok.com> writes:
>
>> Hi Kristoffer,
>>
>> may be it is worth to silently (or at least with rc=0) allow deletion of
>> non-existing or already-deleted configuration statements?
>>
>> Background for that is that I keep track of the all configuration
>> statements myself, and, when I delete some resources (together with
>> accompanying constraints), they may go "out-of-order" to 'crm configure
>> delete', thus some constraints are automatically deleted when deleting
>> "lower" resource before the "upper" one. That leads to the whole crm
>> script to fail.
>
> Hmm, I am not sure about doing this by default, since we would want to
> show some kind of indication that a resource name may have been
> misspelled for example... But I can imagine having a command line flag
> for being more flexible in this regard.

I'd agree: Stopping a stopped resource is different from deleting a resource
that does not exist (maybe never existed). Maybe (for scripting) crm shell
would benefit from a "crm resource exists <rsc_name>" returning a status
(yes/no/error).

>
> I will look at how it works now.
>
> BTW, I suspect that passing the --wait flag to crm while running
> commands in this way may help you. Although I am not sure I entirely
> understand what it is you are doing :)
>
> Cheers,
> Kristoffer
>
>>
>> Best,
>> Vladislav
>>
>> 13.02.2015 17:03, Vladislav Bogdanov wrote:
>>> Hi,
>>>
>>> Following fails with the current crmsh (e4b10ee).
>>> # crm resource stop cl-http-lv
>>> # crm resource stop cl-http-lv
>>> ERROR: crm_diff apparently failed to produce the diff (rc=0)
>>> ERROR: Failed to commit updates to cl-http-lv
>>> # echo $?
>>> 1
>>>
>>>
>>> Best,
>>> Vladislav
>>> _______________________________________________
>>> Linux-HA mailing list
>>> Linux-HA@lists.linux-ha.org
>>> http://lists.linux-ha.org/mailman/listinfo/linux-ha
>>> See also: http://linux-ha.org/ReportingProblems
>>
>>
>
> --
> // Kristoffer Grönlund
> // kgronlund@suse.com
> _______________________________________________
> Linux-HA mailing list
> Linux-HA@lists.linux-ha.org
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems



_______________________________________________
Linux-HA mailing list
Linux-HA@lists.linux-ha.org
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

>>> Lars Marowsky-Bree <lmb@suse.com> schrieb am 16.02.2015 um 12:34 in
Nachricht
<20150216113433.GB4963@suse.de>:
> On 2015-02-16T09:20:22, Kristoffer Grönlund <kgronlund@suse.com> wrote:
>
>> Actually, I decided that it does make sense to return 0 as the error
>> code even if the resource to delete doesn't exist, so I pushed a commit
>> to change this. The error message is still printed, though.
>
> I'm not sure I agree, for once.
>
> Idempotency is for resource agent operations, not necessarily all
> operations everywhere. Especially because crmsh doesn't know whether the
> object doesn't exist because it was deleted, or because it was
> misspelled.

So far I was assuming we are talking about "stopping a stopped resource", not
"stopping a non-existing resource". I thought that crm shell would clearly
distinguish between those. The latter should be an error, of course.


>
> Compare the Unix-as-little-else rm command; rm -f /tmp/idontexist will
> give an error code.

You are wrong:
windl@v04:~> rm -f /tmp/foobar
windl@v04:~> echo $?
0

>
> Now a caller of crmsh has to *parse the output* to know whether the
> delete command succeeded or not, which is rather non-trivial.

Well ,doesn't crm shell parse the CIB all the time? If so it should be able to
find out whether the resource is still there. Anyway now you are talking about
deleting a resource while the subject is on stopping a stopped resource.

>
> If the caller doesn't care whether the command succeeded or not, it
> should be the caller that ignores the error code.

Who is "the caller" for you now? I don't quite get is: The caller of crm
shell, or the crm shell?

>
> Or if you want to get real fancy, return different exit codes for
> "referenced object does not exist", or "generic syntax error".

I Agree.

>
>> >> Following fails with the current crmsh (e4b10ee).
>> >> # crm resource stop cl-http-lv
>> >> # crm resource stop cl-http-lv
>> >> ERROR: crm_diff apparently failed to produce the diff (rc=0)
>> >> ERROR: Failed to commit updates to cl-http-lv
>> >> # echo $?
>> >> 1

I'm not happy with the error messages: Whil technically quite correct, they
are confusing for the user: If the resource doesn't exist, can't crm shell find
out before trying to create/apply a diff? And "apparently failed": Did it, or
didn't it? crm shell should be able to find out!


>
> And, yeah, well, this shouldn't happen. Here idempotency applies ;-)

Sometimes 90% of the code are about things that shouldn't happen; if they do,
it's good to have that code ;-)

Regards,
Ulrich

_______________________________________________
Linux-HA mailing list
Linux-HA@lists.linux-ha.org
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

>>> Kristoffer Grönlund<kgronlund@suse.com> schrieb am 16.02.2015 um 13:12 in
Nachricht <87fva6gim0.fsf@krigpad.kri.gs>:

[...]
> As a sidenote, I find the automatic removal of related constraints
> somewhat strange... To continue the filesystem analogy, it is as if soft

I object: I think this is a _very_ important feature. Somtimes I'd wish "show
changed" could display a word-diff-like output...

> links would be deleted when the file they point to is removed. ;)

Well It's not quite the same: You can recreate the file through the symbolic
link, but you cannot recreate the resource through the constraint.

[...]
Regards,
Ulrich

_______________________________________________
Linux-HA mailing list
Linux-HA@lists.linux-ha.org
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

>>> Dejan Muhamedagic <dejanmm@fastmail.fm> schrieb am 16.02.2015 um 16:08 in
Nachricht <20150216150805.GC27091@tuttle.linbit>:
[...]
> Yes, I see your point and it could just as well be that crmsh is
> really overzealous in this case. Perhaps the removal should be
> deferred until the commit?

I'm confused: Dosn't it? I thought within the "configure" context nothing really happens before "commit"?

[...]

Regards,
Ulrich


_______________________________________________
Linux-HA mailing list
Linux-HA@lists.linux-ha.org
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

18.02.2015 11:03, Ulrich Windl wrote:
>>>> Lars Marowsky-Bree <lmb@suse.com> schrieb am 16.02.2015 um 12:34 in
> Nachricht
> <20150216113433.GB4963@suse.de>:
>> On 2015-02-16T09:20:22, Kristoffer Grönlund <kgronlund@suse.com> wrote:
>>
>>> Actually, I decided that it does make sense to return 0 as the error
>>> code even if the resource to delete doesn't exist, so I pushed a commit
>>> to change this. The error message is still printed, though.
>>
>> I'm not sure I agree, for once.
>>
>> Idempotency is for resource agent operations, not necessarily all
>> operations everywhere. Especially because crmsh doesn't know whether the
>> object doesn't exist because it was deleted, or because it was
>> misspelled.
>
> So far I was assuming we are talking about "stopping a stopped resource", not
> "stopping a non-existing resource". I thought that crm shell would clearly
> distinguish between those. The latter should be an error, of course.

That is my fault, I should be starting the new thread, but I thought
that issue with deleting non-existent object is really very minor.

Best,
Vladislav

_______________________________________________
Linux-HA mailing list
Linux-HA@lists.linux-ha.org
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

Hi,

On Wed, Feb 18, 2015 at 08:21:10AM +0100, Ulrich Windl wrote:
> >>> Kristoffer Grönlund<kgronlund@suse.com> schrieb am 16.02.2015 um 09:15 in
> Nachricht <87oaougtlw.fsf@krigpad.kri.gs>:
> > Vladislav Bogdanov <bubble@hoster-ok.com> writes:
> >
> >> Hi Kristoffer,
> >>
> >> may be it is worth to silently (or at least with rc=0) allow deletion of
> >> non-existing or already-deleted configuration statements?
> >>
> >> Background for that is that I keep track of the all configuration
> >> statements myself, and, when I delete some resources (together with
> >> accompanying constraints), they may go "out-of-order" to 'crm configure
> >> delete', thus some constraints are automatically deleted when deleting
> >> "lower" resource before the "upper" one. That leads to the whole crm
> >> script to fail.
> >
> > Hmm, I am not sure about doing this by default, since we would want to
> > show some kind of indication that a resource name may have been
> > misspelled for example... But I can imagine having a command line flag
> > for being more flexible in this regard.
>
> I'd agree: Stopping a stopped resource is different from deleting a resource
> that does not exist (maybe never existed). Maybe (for scripting) crm shell
> would benefit from a "crm resource exists <rsc_name>" returning a status
> (yes/no/error).

You could use crm resource status for that. It should exit with
error only if the resource doesn't exist.

Thanks,

Dejan

_______________________________________________
Linux-HA mailing list
Linux-HA@lists.linux-ha.org
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

Hi,

On Wed, Feb 18, 2015 at 09:15:57AM +0100, Ulrich Windl wrote:
> >>> Dejan Muhamedagic <dejanmm@fastmail.fm> schrieb am 16.02.2015 um 16:08 in
> Nachricht <20150216150805.GC27091@tuttle.linbit>:
> [...]
> > Yes, I see your point and it could just as well be that crmsh is
> > really overzealous in this case. Perhaps the removal should be
> > deferred until the commit?
>
> I'm confused: Dosn't it? I thought within the "configure" context nothing really happens before "commit"?

crmsh keeps a work copy of the CIB in memory. All configure
commands affect that one. On commit, the memory CIB is copied to
the target CIB, which is typically the live cluster CIB (but it
can also be a shadow CIB, which is just a file).

What I referred to above was the work copy, i.e. that the
"hanging" constraints removal should happen at the commit time,
so that the user has a chance to update them beforehand.

Thanks,

Dejan

>
> [...]
>
> Regards,
> Ulrich
>
>
> _______________________________________________
> Linux-HA mailing list
> Linux-HA@lists.linux-ha.org
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems
_______________________________________________
Linux-HA mailing list
Linux-HA@lists.linux-ha.org
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems