From: "Laszlo Hornyak" <lhornyak(a)redhat.com&gt; To: "engine-devel" <engine-devel(a)ovirt.org&gt;, users(a)ovirt.org Sent: Monday, February 13, 2012 12:32:34 PM Subject: [Users] Autorecovery feature plan for review Hi, Please review the plan document for autorecovery. http://www.ovirt.org/wiki/Features/Autorecovery Thank you, Laszlo _______________________________________________ Users mailing list Users(a)ovirt.org http://lists.ovirt.org/mailman/listinfo/users

Hi, Please review the plan document for autorecovery. http://www.ovirt.org/wiki/Features/Autorecovery

On 14/02/12 05:56, Itamar Heim wrote: > On 02/13/2012 12:32 PM, Laszlo Hornyak wrote: >> Hi, >> >> Please review the plan document for autorecovery. >> http://www.ovirt.org/wiki/Features/Autorecovery > > why would we disable auto recovery by default? it sounds like the > preferred behavior? > I think that by default Laszlo meant in the upgrade process to maintain current behavior. I agree that for new entities the default should be true.

From: "Yair Zaslavsky" <yzaslavs(a)redhat.com&gt; To: engine-devel(a)ovirt.org Sent: Tuesday, February 14, 2012 9:20:10 AM Subject: Re: [Engine-devel] Autorecovery feature plan for review On 02/14/2012 08:59 AM, Itamar Heim wrote: > On 02/14/2012 08:57 AM, Livnat Peer wrote: >> On 14/02/12 05:56, Itamar Heim wrote: >>> On 02/13/2012 12:32 PM, Laszlo Hornyak wrote: >>>> Hi, >>>> >>>> Please review the plan document for autorecovery. >>>> http://www.ovirt.org/wiki/Features/Autorecovery >>> >>> why would we disable auto recovery by default? it sounds like the >>> preferred behavior? >>> >> >> I think that by default Laszlo meant in the upgrade process to >> maintain >> current behavior. >> >> I agree that for new entities the default should be true. > > i think the only combination which will allow this is for db to > default > to false and code to default to true for this property? Why can't we during upgrade process set to all existing entities in DB the value to false, but still have the column defined as "default true"?

> _______________________________________________ > Engine-devel mailing list > Engine-devel(a)ovirt.org > http://lists.ovirt.org/mailman/listinfo/engine-devel _______________________________________________ Engine-devel mailing list Engine-devel(a)ovirt.org http://lists.ovirt.org/mailman/listinfo/engine-devel

On 02/14/2012 09:48 AM, Omer Frenkel wrote: > > ----- Original Message ----- >> From: "Yair Zaslavsky&quot;&lt;yzaslavs(a)redhat.com&gt; >> To: engine-devel(a)ovirt.org >> Sent: Tuesday, February 14, 2012 9:20:10 AM >> Subject: Re: [Engine-devel] Autorecovery feature plan for review >> >> On 02/14/2012 08:59 AM, Itamar Heim wrote: >>> On 02/14/2012 08:57 AM, Livnat Peer wrote: >>>> On 14/02/12 05:56, Itamar Heim wrote: >>>>> On 02/13/2012 12:32 PM, Laszlo Hornyak wrote: >>>>>> Hi, >>>>>> >>>>>> Please review the plan document for autorecovery. >>>>>> http://www.ovirt.org/wiki/Features/Autorecovery >>>>> why would we disable auto recovery by default? it sounds like the >>>>> preferred behavior? >>>>> >>>> I think that by default Laszlo meant in the upgrade process to >>>> maintain >>>> current behavior.

>>>> >>>> I agree that for new entities the default should be true. >>> i think the only combination which will allow this is for db to >>> default >>> to false and code to default to true for this property? >> Why can't we during upgrade process set to all existing entities in >> DB >> the value to false, but still have the column defined as "default >> true"? > why all the trouble? i think this field should be mandatory as any other field, > user has to specify it during the entity creation, right where he provide the name and any other field for the new entity. Fine by me.

>>> _______________________________________________ >>> Engine-devel mailing list >>> Engine-devel(a)ovirt.org >>> http://lists.ovirt.org/mailman/listinfo/engine-devel >> _______________________________________________ >> Engine-devel mailing list >> Engine-devel(a)ovirt.org >> http://lists.ovirt.org/mailman/listinfo/engine-devel >> _______________________________________________ Engine-devel mailing list Engine-devel(a)ovirt.org http://lists.ovirt.org/mailman/listinfo/engine-devel

From: "Livnat Peer" <lpeer(a)redhat.com&gt; To: "Yaniv Kaul" <ykaul(a)redhat.com&gt; Cc: engine-devel(a)ovirt.org Sent: Tuesday, February 14, 2012 1:00:35 PM Subject: Re: [Engine-devel] Autorecovery feature plan for review On 14/02/12 11:45, Yaniv Kaul wrote: > On 02/14/2012 09:59 AM, Yair Zaslavsky wrote: >> On 02/14/2012 09:48 AM, Omer Frenkel wrote: >>> >>> ----- Original Message ----- >>>> From: "Yair Zaslavsky&quot;&lt;yzaslavs(a)redhat.com&gt; >>>> To: engine-devel(a)ovirt.org >>>> Sent: Tuesday, February 14, 2012 9:20:10 AM >>>> Subject: Re: [Engine-devel] Autorecovery feature plan for review >>>> >>>> On 02/14/2012 08:59 AM, Itamar Heim wrote: >>>>> On 02/14/2012 08:57 AM, Livnat Peer wrote: >>>>>> On 14/02/12 05:56, Itamar Heim wrote: >>>>>>> On 02/13/2012 12:32 PM, Laszlo Hornyak wrote: >>>>>>>> Hi, >>>>>>>> >>>>>>>> Please review the plan document for autorecovery. >>>>>>>> http://www.ovirt.org/wiki/Features/Autorecovery >>>>>>> why would we disable auto recovery by default? it sounds like >>>>>>> the >>>>>>> preferred behavior? >>>>>>> >>>>>> I think that by default Laszlo meant in the upgrade process to >>>>>> maintain >>>>>> current behavior. > > Why? Why not improve their user experience and provide them with > such > feature? Current behaviour sucks - as your system admin. > I don't have objections either way. Laszlo - let's update the wiki to upgrade by default to true, if we'll get good reason why not to upgrade to true then we can open it again for discussion.

>>>>>> >>>>>> I agree that for new entities the default should be true. >>>>> i think the only combination which will allow this is for db to >>>>> default >>>>> to false and code to default to true for this property? >>>> Why can't we during upgrade process set to all existing entities >>>> in >>>> DB >>>> the value to false, but still have the column defined as >>>> "default >>>> true"? >>> why all the trouble? i think this field should be mandatory as >>> any >>> other field, >>> user has to specify it during the entity creation, right where he >>> provide the name and any other field for the new entity. >> Fine by me. > > I'm not sure I see the reason a user would want to turn it off, on > a > per-object basis. > If it's in the 'Advanced' settings of a host/storage, fine, but > otherwise, it's just another cryptic feature to turn on/off. > Y. > >> >>>>> _______________________________________________ >>>>> Engine-devel mailing list >>>>> Engine-devel(a)ovirt.org >>>>> http://lists.ovirt.org/mailman/listinfo/engine-devel >>>> _______________________________________________ >>>> Engine-devel mailing list >>>> Engine-devel(a)ovirt.org >>>> http://lists.ovirt.org/mailman/listinfo/engine-devel >>>> >> _______________________________________________ >> Engine-devel mailing list >> Engine-devel(a)ovirt.org >> http://lists.ovirt.org/mailman/listinfo/engine-devel > > _______________________________________________ > Engine-devel mailing list > Engine-devel(a)ovirt.org > http://lists.ovirt.org/mailman/listinfo/engine-devel _______________________________________________ Engine-devel mailing list Engine-devel(a)ovirt.org http://lists.ovirt.org/mailman/listinfo/engine-devel

On 02/14/2012 08:59 AM, Itamar Heim wrote: > On 02/14/2012 08:57 AM, Livnat Peer wrote: >> On 14/02/12 05:56, Itamar Heim wrote: >>> On 02/13/2012 12:32 PM, Laszlo Hornyak wrote: >>>> Hi, >>>> >>>> Please review the plan document for autorecovery. >>>> http://www.ovirt.org/wiki/Features/Autorecovery >>> >>> why would we disable auto recovery by default? it sounds like the >>> preferred behavior? >>> >> >> I think that by default Laszlo meant in the upgrade process to maintain >> current behavior. >> >> I agree that for new entities the default should be true. > > i think the only combination which will allow this is for db to default > to false and code to default to true for this property? Why can't we during upgrade process set to all existing entities in DB the value to false, but still have the column defined as "default true"?

On 02/14/2012 10:03 PM, Itamar Heim wrote: > On 02/14/2012 09:20 AM, Yair Zaslavsky wrote: >> On 02/14/2012 08:59 AM, Itamar Heim wrote: >>> On 02/14/2012 08:57 AM, Livnat Peer wrote: >>>> On 14/02/12 05:56, Itamar Heim wrote: >>>>> On 02/13/2012 12:32 PM, Laszlo Hornyak wrote: >>>>>> Hi, >>>>>> >>>>>> Please review the plan document for autorecovery. >>>>>> http://www.ovirt.org/wiki/Features/Autorecovery >>>>> >>>>> why would we disable auto recovery by default? it sounds like >>>>> the >>>>> preferred behavior? >>>>> >>>> >>>> I think that by default Laszlo meant in the upgrade process to >>>> maintain >>>> current behavior. >>>> >>>> I agree that for new entities the default should be true. >>> >>> i think the only combination which will allow this is for db to >>> default >>> to false and code to default to true for this property? >> Why can't we during upgrade process set to all existing entities >> in DB >> the value to false, but still have the column defined as "default >> true"? > > because upgrade and clean install are running the same scripts? I guess I still fail to understand. Scenarios (as both upgrade and clean install run the same scripts) a. In environment to be upgraded we have X entities that are non recoverable - after upgrade these X entities have the boolean flag set to false. New entities in the system will be created with auto recoverable set to true. b. In environment to be clean installed -we have 0 existing entities - after clean install all new entities in the system will be create with auto recoverable set to true. Will this be considered a bad behavior?

_______________________________________________ Engine-devel mailing list Engine-devel(a)ovirt.org http://lists.ovirt.org/mailman/listinfo/engine-devel

On 02/14/2012 10:03 PM, Itamar Heim wrote: > On 02/14/2012 09:20 AM, Yair Zaslavsky wrote: >> On 02/14/2012 08:59 AM, Itamar Heim wrote: >>> On 02/14/2012 08:57 AM, Livnat Peer wrote: >>>> On 14/02/12 05:56, Itamar Heim wrote: >>>>> On 02/13/2012 12:32 PM, Laszlo Hornyak wrote: >>>>>> Hi, >>>>>> >>>>>> Please review the plan document for autorecovery. >>>>>> http://www.ovirt.org/wiki/Features/Autorecovery >>>>> why would we disable auto recovery by default? it sounds like the >>>>> preferred behavior? >>>>> >>>> I think that by default Laszlo meant in the upgrade process to maintain >>>> current behavior. >>>> >>>> I agree that for new entities the default should be true. >>> i think the only combination which will allow this is for db to default >>> to false and code to default to true for this property? >> Why can't we during upgrade process set to all existing entities in DB >> the value to false, but still have the column defined as "default true"? > because upgrade and clean install are running the same scripts? I guess I still fail to understand. Scenarios (as both upgrade and clean install run the same scripts) a. In environment to be upgraded we have X entities that are non recoverable - after upgrade these X entities have the boolean flag set to false. New entities in the system will be created with auto recoverable set to true.

b. In environment to be clean installed -we have 0 existing entities - after clean install all new entities in the system will be create with auto recoverable set to true. Will this be considered a bad behavior? _______________________________________________ Engine-devel mailing list Engine-devel(a)ovirt.org http://lists.ovirt.org/mailman/listinfo/engine-devel

From: "Ayal Baron" <abaron(a)redhat.com&gt; To: "Yaniv Kaul" <ykaul(a)redhat.com&gt; Cc: engine-devel(a)ovirt.org Sent: Wednesday, February 15, 2012 12:19:48 PM Subject: Re: [Engine-devel] Autorecovery feature plan for review > > I still fail to understand why you 'punish' existing objects and > not > giving them the new feature enabled by default. This is not a feature, it's a bug!

This should not be treated as a feature and this should not be configurable!

Today an object moves to non-operational due to state reported by vdsm. The object should immediately return to up the moment vdsm reports the object as ok (this means that you don't stop monitoring just because there is an error). That's it. no db field and no nothing... This pertains to storage domains, network, host status, whatever. > Y. > > > b. In environment to be clean installed -we have 0 existing > > entities - > > after clean install all new entities in the system will be create > > with > > auto recoverable set to true. > > Will this be considered a bad behavior? > > > > > > _______________________________________________ > > Engine-devel mailing list > > Engine-devel(a)ovirt.org > > http://lists.ovirt.org/mailman/listinfo/engine-devel > > _______________________________________________ > Engine-devel mailing list > Engine-devel(a)ovirt.org > http://lists.ovirt.org/mailman/listinfo/engine-devel > _______________________________________________ Engine-devel mailing list Engine-devel(a)ovirt.org http://lists.ovirt.org/mailman/listinfo/engine-devel

From: "Ayal Baron" <abaron(a)redhat.com&gt; To: "Laszlo Hornyak" <lhornyak(a)redhat.com&gt; Cc: engine-devel(a)ovirt.org, "Yaniv Kaul" <ykaul(a)redhat.com&gt; Sent: Wednesday, February 15, 2012 12:46:05 PM Subject: Re: [Engine-devel] Autorecovery feature plan for review ----- Original Message ----- > Hi Ayal, > > ----- Original Message ----- > > From: "Ayal Baron" <abaron(a)redhat.com&gt; > > To: "Yaniv Kaul" <ykaul(a)redhat.com&gt; > > Cc: engine-devel(a)ovirt.org > > Sent: Wednesday, February 15, 2012 12:19:48 PM > > Subject: Re: [Engine-devel] Autorecovery feature plan for review > > > > > > > > > > I still fail to understand why you 'punish' existing objects > > > and > > > not > > > giving them the new feature enabled by default. > > > > This is not a feature, it's a bug! > > Whatever we call it, it is a change in behavior. We agreed that it > will be enabled for all existing objects by default. > > http://globalnerdy.com/wordpress/wp-content/uploads/2007/12/bug_vs_featur... > > > This should not be treated as a feature and this should not be > > configurable! > > I can imagine some situations when I would not like the > autorecovery > to happen, but if everyone agrees not to make it configurable, I > will just remove it from my patchset. It's not autorecovery, you're not recovering anything. You're reflecting the fact that the resource is back to normal (not due to anything that the engine did). This is why it is a bug today. This is why it should not be configurable. > > > Today an object moves to non-operational due to state reported by > > vdsm. The object should immediately return to up the moment vdsm > > reports the object as ok (this means that you don't stop > > monitoring > > just because there is an error). > > That's it. no db field and no nothing... > > This pertains to storage domains, network, host status, whatever. > > > > > Y. > > > > > > > b. In environment to be clean installed -we have 0 existing > > > > entities - > > > > after clean install all new entities in the system will be > > > > create > > > > with > > > > auto recoverable set to true. > > > > Will this be considered a bad behavior? > > > > > > > > > > > > _______________________________________________ > > > > Engine-devel mailing list > > > > Engine-devel(a)ovirt.org > > > > http://lists.ovirt.org/mailman/listinfo/engine-devel > > > > > > _______________________________________________ > > > Engine-devel mailing list > > > Engine-devel(a)ovirt.org > > > http://lists.ovirt.org/mailman/listinfo/engine-devel > > > > > _______________________________________________ > > Engine-devel mailing list > > Engine-devel(a)ovirt.org > > http://lists.ovirt.org/mailman/listinfo/engine-devel > > >

----- Original Message ----- > Hi, > > A short summary from the call today, please correct me if I forgot or > misunderstood something. > > Ayal argued that the failed host/storagedomain should be reactivated > by a periodically executed job, he would prefer if the engine could > [try to] correct the problem right on discovery. > Livnat's point was that this is hard to implement and it is OK if we > move it to Nonoperational state and periodically check it again. > > There was a little arguing if we call the current behavior a bug or a > missing behavior, I believe this is not quite important. > > I did not fully understand the last few sentences from Livant, did we > manage to agree in a change in the plan? A couple of points that we agreed upon: 1. no need for new mechanism, just initiate this from the monitoring context. Preferably, if not difficult, evaluate the monitoring data, if host should remain in non-op then don't bother running initVdsOnUp 2. configuration of when to call initvdsonup is orthogonal to auto-init behaviour and if introduced should be on by default and user should be able to configure this either on or off for the host in general (no lower granularity) and can only be configured via the API. When disabled initVdsOnUp would be called only when admin activates the host/storage and any error would keep it inactive (I still don't understand why this is at all needed but whatever).

Note that going forward what I envision is engine pushing down the entire host configuration once and from that point on the host would try to keep this configuration up and running. Once this happens there will be no need for initVdsOnUp at all. > > Anyway, I agree with Ayal that it would be very nice if the engine > could fix the issues right on discovery, but I also agree that this > feature would take a bigger effort. It would be nice to know what > effort it would take to get the monitoring do this safely. Could we > still call it monitoring then? >

> Laszlo > > ----- Original Message ----- >> From: "Ayal Baron" <abaron(a)redhat.com&gt; >> To: "Laszlo Hornyak" <lhornyak(a)redhat.com&gt; >> Cc: engine-devel(a)ovirt.org, "Yaniv Kaul" <ykaul(a)redhat.com&gt; >> Sent: Wednesday, February 15, 2012 12:46:05 PM >> Subject: Re: [Engine-devel] Autorecovery feature plan for review >> >> >> >> ----- Original Message ----- >>> Hi Ayal, >>> >>> ----- Original Message ----- >>>> From: "Ayal Baron" <abaron(a)redhat.com&gt; >>>> To: "Yaniv Kaul" <ykaul(a)redhat.com&gt; >>>> Cc: engine-devel(a)ovirt.org >>>> Sent: Wednesday, February 15, 2012 12:19:48 PM >>>> Subject: Re: [Engine-devel] Autorecovery feature plan for >>>> review >>>> >>>> >>>>> >>>>> I still fail to understand why you 'punish' existing objects >>>>> and >>>>> not >>>>> giving them the new feature enabled by default. >>>> >>>> This is not a feature, it's a bug! >>> >>> Whatever we call it, it is a change in behavior. We agreed that >>> it >>> will be enabled for all existing objects by default. >>> >>> http://globalnerdy.com/wordpress/wp-content/uploads/2007/12/bug_vs_featur... >>> >>>> This should not be treated as a feature and this should not be >>>> configurable! >>> >>> I can imagine some situations when I would not like the >>> autorecovery >>> to happen, but if everyone agrees not to make it configurable, I >>> will just remove it from my patchset. >> >> It's not autorecovery, you're not recovering anything. You're >> reflecting the fact that the resource is back to normal (not due to >> anything that the engine did). >> This is why it is a bug today. >> This is why it should not be configurable. >> >>> >>>> Today an object moves to non-operational due to state reported >>>> by >>>> vdsm. The object should immediately return to up the moment >>>> vdsm >>>> reports the object as ok (this means that you don't stop >>>> monitoring >>>> just because there is an error). >>>> That's it. no db field and no nothing... >>>> This pertains to storage domains, network, host status, >>>> whatever. >>>> >>>>> Y. >>>>> >>>>>> b. In environment to be clean installed -we have 0 existing >>>>>> entities - >>>>>> after clean install all new entities in the system will be >>>>>> create >>>>>> with >>>>>> auto recoverable set to true. >>>>>> Will this be considered a bad behavior? >>>>>> >>>>>> >>>>>> _______________________________________________ >>>>>> Engine-devel mailing list >>>>>> Engine-devel(a)ovirt.org >>>>>> http://lists.ovirt.org/mailman/listinfo/engine-devel >>>>> >>>>> _______________________________________________ >>>>> Engine-devel mailing list >>>>> Engine-devel(a)ovirt.org >>>>> http://lists.ovirt.org/mailman/listinfo/engine-devel >>>>> >>>> _______________________________________________ >>>> Engine-devel mailing list >>>> Engine-devel(a)ovirt.org >>>> http://lists.ovirt.org/mailman/listinfo/engine-devel >>>> >>> >> > _______________________________________________ Engine-devel mailing list Engine-devel(a)ovirt.org http://lists.ovirt.org/mailman/listinfo/engine-devel

On 02/15/2012 07:02 PM, Livnat Peer wrote: > On 15/02/12 18:28, Ayal Baron wrote: >> >> >> ----- Original Message ----- >>> Hi, >>> >>> A short summary from the call today, please correct me if I forgot or >>> misunderstood something. >>> >>> Ayal argued that the failed host/storagedomain should be reactivated >>> by a periodically executed job, he would prefer if the engine could >>> [try to] correct the problem right on discovery. >>> Livnat's point was that this is hard to implement and it is OK if we >>> move it to Nonoperational state and periodically check it again. >>> >>> There was a little arguing if we call the current behavior a bug or a >>> missing behavior, I believe this is not quite important. >>> >>> I did not fully understand the last few sentences from Livant, did we >>> manage to agree in a change in the plan? >> >> A couple of points that we agreed upon: >> 1. no need for new mechanism, just initiate this from the monitoring >> context. >> Preferably, if not difficult, evaluate the monitoring data, if >> host should remain in non-op then don't bother running initVdsOnUp >> 2. configuration of when to call initvdsonup is orthogonal to >> auto-init behaviour and if introduced should be on by default and >> user should be able to configure this either on or off for the host >> in general (no lower granularity) and can only be configured via the >> API. >> When disabled initVdsOnUp would be called only when admin activates >> the host/storage and any error would keep it inactive (I still don't >> understand why this is at all needed but whatever). >> > > Also a note from Moran on the call was to check if we can unify the > non-operational and Error statuses of the host. > It was mentioned on the call that the reason for having ERROR state is > for recovery (time out of the error state) but since we are about to > recover from non-operational status as well there is no reason to have > two different statuses. they are not exactly the same. or should i say, error is supposed to be when reason isn't related to host being non-operational. what is error state? a host will go into error state if it fails to run 3 (configurable) VMs, that succeeded running on other host on retry. i.e., something is wrong with that host, failing to launch VMs. as it happens, it already "auto recovers" for this mode after a certain period of time. why? because the host will fail to run virtual machines, and will be the least loaded, so it will be the first target selected to run them, which will continue to fail. so there is a negative scoring mechanism on number of errors, till host is taken out for a while. (I don't remember if the reverse is true and the VM goes into error mode if the VM failed to launch on all hosts per number of retries. i think this wasn't needed and user just got an error in audit log) i can see two reasons a host will go into error state: 1. monitoring didn't detect an issue yet, and host would have/will/should go into non-operational mode. if host will go into non-operational mode, and will auto recover with the above flow, i guess it is fine. 2. cause for failure isn't something we monitor for (upgraded to a bad version of qemu, or qemu got corrupted). now, the error mode was developed quite a long time ago (august 2007 iirc), so could be it mostly compensated for the first reason which is now better monitored. i wonder how often error state is seen due to a reason which isn't monitored already. moran - do you have examples of when you see error state of hosts?

On 02/16/2012 09:29 AM, Moran Goldboim wrote: > On 02/16/2012 12:38 AM, Itamar Heim wrote: >> On 02/15/2012 07:02 PM, Livnat Peer wrote: >>> On 15/02/12 18:28, Ayal Baron wrote: >>>> >>>> >>>> ----- Original Message ----- >>>>> Hi, >>>>> >>>>> A short summary from the call today, please correct me if I >>>>> forgot or >>>>> misunderstood something. >>>>> >>>>> Ayal argued that the failed host/storagedomain should be reactivated >>>>> by a periodically executed job, he would prefer if the engine could >>>>> [try to] correct the problem right on discovery. >>>>> Livnat's point was that this is hard to implement and it is OK if we >>>>> move it to Nonoperational state and periodically check it again. >>>>> >>>>> There was a little arguing if we call the current behavior a bug >>>>> or a >>>>> missing behavior, I believe this is not quite important. >>>>> >>>>> I did not fully understand the last few sentences from Livant, >>>>> did we >>>>> manage to agree in a change in the plan? >>>> >>>> A couple of points that we agreed upon: >>>> 1. no need for new mechanism, just initiate this from the >>>> monitoring context. >>>> Preferably, if not difficult, evaluate the monitoring data, if >>>> host should remain in non-op then don't bother running initVdsOnUp >>>> 2. configuration of when to call initvdsonup is orthogonal to >>>> auto-init behaviour and if introduced should be on by default and >>>> user should be able to configure this either on or off for the >>>> host in general (no lower granularity) and can only be configured >>>> via the API. >>>> When disabled initVdsOnUp would be called only when admin >>>> activates the host/storage and any error would keep it inactive (I >>>> still don't understand why this is at all needed but whatever). >>>> >>> >>> Also a note from Moran on the call was to check if we can unify the >>> non-operational and Error statuses of the host. >>> It was mentioned on the call that the reason for having ERROR state is >>> for recovery (time out of the error state) but since we are about to >>> recover from non-operational status as well there is no reason to have >>> two different statuses. >> >> they are not exactly the same. >> or should i say, error is supposed to be when reason isn't related >> to host being non-operational. >> >> what is error state? >> a host will go into error state if it fails to run 3 (configurable) >> VMs, that succeeded running on other host on retry. >> i.e., something is wrong with that host, failing to launch VMs. >> as it happens, it already "auto recovers" for this mode after a >> certain period of time. >> >> why? because the host will fail to run virtual machines, and will be >> the least loaded, so it will be the first target selected to run >> them, which will continue to fail. >> >> so there is a negative scoring mechanism on number of errors, till >> host is taken out for a while. >> >> (I don't remember if the reverse is true and the VM goes into error >> mode if the VM failed to launch on all hosts per number of retries. >> i think this wasn't needed and user just got an error in audit log) >> >> i can see two reasons a host will go into error state: >> 1. monitoring didn't detect an issue yet, and host would >> have/will/should go into non-operational mode. >> if host will go into non-operational mode, and will auto recover >> with the above flow, i guess it is fine. >> >> 2. cause for failure isn't something we monitor for (upgraded to a >> bad version of qemu, or qemu got corrupted). >> >> now, the error mode was developed quite a long time ago (august 2007 >> iirc), so could be it mostly compensated for the first reason which >> is now better monitored. >> i wonder how often error state is seen due to a reason which isn't >> monitored already. >> moran - do you have examples of when you see error state of hosts? > > usually it happened when there were a problematic/ misconfigurated > vdsm / libvirt which failed to run vms (nothing we can recover from)- > i haven't faced the issue of "host it too loaded" that status has > some other syndromes, however the behaviour on that state is very > much the same -waiting for 30 min (?) and than move it to activated. > Moran. 'host is too loaded' is too loaded is the only transient state where a temporary 'error' state makes sense, but in the same time, it can also fit the 'non operational' state description. From my experience, the problem with KVM/libvirt/VDSM mis-configured is never temporary, (= magically solved by itself, without concrete user intervention). IMHO, it should move the host to an error state that would not automatically recover from. Regardless, consolidating the names of the states ('inactive, detached, non operational, maintenance, error, unknown' ...) would be nice too. Probably can't be done for all, of course. Y.

> _______________________________________________ > Engine-devel mailing list > Engine-devel(a)ovirt.org > http://lists.ovirt.org/mailman/listinfo/engine-devel

----- Original Message ----- > From: "Moran Goldboim&quot;&lt;mgoldboi(a)redhat.com&gt; > To: "Yaniv Kaul&quot;&lt;ykaul(a)redhat.com&gt; > Cc: engine-devel(a)ovirt.org > Sent: Thursday, February 16, 2012 10:01:37 AM > Subject: Re: [Engine-devel] Autorecovery feature plan for review > > On 02/16/2012 09:35 AM, Yaniv Kaul wrote: >> On 02/16/2012 09:29 AM, Moran Goldboim wrote: >>> On 02/16/2012 12:38 AM, Itamar Heim wrote: >>>> On 02/15/2012 07:02 PM, Livnat Peer wrote: >>>>> On 15/02/12 18:28, Ayal Baron wrote: >>>>>> >>>>>> ----- Original Message ----- >>>>>>> Hi, >>>>>>> >>>>>>> A short summary from the call today, please correct me if I >>>>>>> forgot or >>>>>>> misunderstood something. >>>>>>> >>>>>>> Ayal argued that the failed host/storagedomain should be >>>>>>> reactivated >>>>>>> by a periodically executed job, he would prefer if the engine >>>>>>> could >>>>>>> [try to] correct the problem right on discovery. >>>>>>> Livnat's point was that this is hard to implement and it is OK >>>>>>> if we >>>>>>> move it to Nonoperational state and periodically check it >>>>>>> again. >>>>>>> >>>>>>> There was a little arguing if we call the current behavior a >>>>>>> bug >>>>>>> or a >>>>>>> missing behavior, I believe this is not quite important. >>>>>>> >>>>>>> I did not fully understand the last few sentences from Livant, >>>>>>> did we >>>>>>> manage to agree in a change in the plan? >>>>>> A couple of points that we agreed upon: >>>>>> 1. no need for new mechanism, just initiate this from the >>>>>> monitoring context. >>>>>> Preferably, if not difficult, evaluate the monitoring data, >>>>>> if >>>>>> host should remain in non-op then don't bother running >>>>>> initVdsOnUp >>>>>> 2. configuration of when to call initvdsonup is orthogonal to >>>>>> auto-init behaviour and if introduced should be on by default >>>>>> and >>>>>> user should be able to configure this either on or off for the >>>>>> host in general (no lower granularity) and can only be >>>>>> configured >>>>>> via the API. >>>>>> When disabled initVdsOnUp would be called only when admin >>>>>> activates the host/storage and any error would keep it inactive >>>>>> (I >>>>>> still don't understand why this is at all needed but whatever). >>>>>> >>>>> Also a note from Moran on the call was to check if we can unify >>>>> the >>>>> non-operational and Error statuses of the host. >>>>> It was mentioned on the call that the reason for having ERROR >>>>> state is >>>>> for recovery (time out of the error state) but since we are >>>>> about to >>>>> recover from non-operational status as well there is no reason >>>>> to have >>>>> two different statuses. >>>> they are not exactly the same. >>>> or should i say, error is supposed to be when reason isn't >>>> related >>>> to host being non-operational. >>>> >>>> what is error state? >>>> a host will go into error state if it fails to run 3 >>>> (configurable) >>>> VMs, that succeeded running on other host on retry. >>>> i.e., something is wrong with that host, failing to launch VMs. >>>> as it happens, it already "auto recovers" for this mode after a >>>> certain period of time. >>>> >>>> why? because the host will fail to run virtual machines, and will >>>> be >>>> the least loaded, so it will be the first target selected to run >>>> them, which will continue to fail. >>>> >>>> so there is a negative scoring mechanism on number of errors, >>>> till >>>> host is taken out for a while. >>>> >>>> (I don't remember if the reverse is true and the VM goes into >>>> error >>>> mode if the VM failed to launch on all hosts per number of >>>> retries. >>>> i think this wasn't needed and user just got an error in audit >>>> log) >>>> >>>> i can see two reasons a host will go into error state: >>>> 1. monitoring didn't detect an issue yet, and host would >>>> have/will/should go into non-operational mode. >>>> if host will go into non-operational mode, and will auto recover >>>> with the above flow, i guess it is fine. >>>> >>>> 2. cause for failure isn't something we monitor for (upgraded to >>>> a >>>> bad version of qemu, or qemu got corrupted). >>>> >>>> now, the error mode was developed quite a long time ago (august >>>> 2007 >>>> iirc), so could be it mostly compensated for the first reason >>>> which >>>> is now better monitored. >>>> i wonder how often error state is seen due to a reason which >>>> isn't >>>> monitored already. >>>> moran - do you have examples of when you see error state of >>>> hosts? >>> usually it happened when there were a problematic/ misconfigurated >>> vdsm / libvirt which failed to run vms (nothing we can recover >>> from)- >>> i haven't faced the issue of "host it too loaded" that status has >>> some other syndromes, however the behaviour on that state is very >>> much the same -waiting for 30 min (?) and than move it to >>> activated. >>> Moran. >> 'host is too loaded' is too loaded is the only transient state >> where a >> temporary 'error' state makes sense, but in the same time, it can >> also >> fit the 'non operational' state description. >> From my experience, the problem with KVM/libvirt/VDSM >> mis-configured >> is never temporary, (= magically solved by itself, without concrete >> user intervention). IMHO, it should move the host to an error state >> that would not automatically recover from. >> Regardless, consolidating the names of the states ('inactive, >> detached, non operational, maintenance, error, unknown' ...) would >> be >> nice too. Probably can't be done for all, of course. >> Y. > agreed, most of the causes of ERROR state aren't transient, but looks > to > me as if this state is redundant and could be taken care as part of > the > other host states, since the way it's being used today isn't very > helpful as well. > Moran. However, I can envision an ERROR state that you don't want to keep retry mechanism on... which might be a different behavior than the NON-OP one.

> >> >>> _______________________________________________ >>> Engine-devel mailing list >>> Engine-devel(a)ovirt.org >>> http://lists.ovirt.org/mailman/listinfo/engine-devel > _______________________________________________ > Engine-devel mailing list > Engine-devel(a)ovirt.org > http://lists.ovirt.org/mailman/listinfo/engine-devel >

On 16/02/12 10:01, Moran Goldboim wrote: > On 02/16/2012 09:35 AM, Yaniv Kaul wrote: >> On 02/16/2012 09:29 AM, Moran Goldboim wrote: >>> On 02/16/2012 12:38 AM, Itamar Heim wrote: >>>> On 02/15/2012 07:02 PM, Livnat Peer wrote: >>>>> On 15/02/12 18:28, Ayal Baron wrote: >>>>>> >>>>>> >>>>>> ----- Original Message ----- >>>>>>> Hi, >>>>>>> >>>>>>> A short summary from the call today, please correct me if I >>>>>>> forgot or >>>>>>> misunderstood something. >>>>>>> >>>>>>> Ayal argued that the failed host/storagedomain should be reactivated >>>>>>> by a periodically executed job, he would prefer if the engine could >>>>>>> [try to] correct the problem right on discovery. >>>>>>> Livnat's point was that this is hard to implement and it is OK if we >>>>>>> move it to Nonoperational state and periodically check it again. >>>>>>> >>>>>>> There was a little arguing if we call the current behavior a bug >>>>>>> or a >>>>>>> missing behavior, I believe this is not quite important. >>>>>>> >>>>>>> I did not fully understand the last few sentences from Livant, >>>>>>> did we >>>>>>> manage to agree in a change in the plan? >>>>>> >>>>>> A couple of points that we agreed upon: >>>>>> 1. no need for new mechanism, just initiate this from the >>>>>> monitoring context. >>>>>> Preferably, if not difficult, evaluate the monitoring data, if >>>>>> host should remain in non-op then don't bother running initVdsOnUp >>>>>> 2. configuration of when to call initvdsonup is orthogonal to >>>>>> auto-init behaviour and if introduced should be on by default and >>>>>> user should be able to configure this either on or off for the >>>>>> host in general (no lower granularity) and can only be configured >>>>>> via the API. >>>>>> When disabled initVdsOnUp would be called only when admin >>>>>> activates the host/storage and any error would keep it inactive (I >>>>>> still don't understand why this is at all needed but whatever). >>>>>> >>>>> >>>>> Also a note from Moran on the call was to check if we can unify the >>>>> non-operational and Error statuses of the host. >>>>> It was mentioned on the call that the reason for having ERROR state is >>>>> for recovery (time out of the error state) but since we are about to >>>>> recover from non-operational status as well there is no reason to have >>>>> two different statuses. >>>> >>>> they are not exactly the same. >>>> or should i say, error is supposed to be when reason isn't related >>>> to host being non-operational. >>>> >>>> what is error state? >>>> a host will go into error state if it fails to run 3 (configurable) >>>> VMs, that succeeded running on other host on retry. >>>> i.e., something is wrong with that host, failing to launch VMs. >>>> as it happens, it already "auto recovers" for this mode after a >>>> certain period of time. >>>> >>>> why? because the host will fail to run virtual machines, and will be >>>> the least loaded, so it will be the first target selected to run >>>> them, which will continue to fail. >>>> >>>> so there is a negative scoring mechanism on number of errors, till >>>> host is taken out for a while. >>>> >>>> (I don't remember if the reverse is true and the VM goes into error >>>> mode if the VM failed to launch on all hosts per number of retries. >>>> i think this wasn't needed and user just got an error in audit log) >>>> >>>> i can see two reasons a host will go into error state: >>>> 1. monitoring didn't detect an issue yet, and host would >>>> have/will/should go into non-operational mode. >>>> if host will go into non-operational mode, and will auto recover >>>> with the above flow, i guess it is fine. >>>> >>>> 2. cause for failure isn't something we monitor for (upgraded to a >>>> bad version of qemu, or qemu got corrupted). >>>> >>>> now, the error mode was developed quite a long time ago (august 2007 >>>> iirc), so could be it mostly compensated for the first reason which >>>> is now better monitored. >>>> i wonder how often error state is seen due to a reason which isn't >>>> monitored already. >>>> moran - do you have examples of when you see error state of hosts? >>> >>> usually it happened when there were a problematic/ misconfigurated >>> vdsm / libvirt which failed to run vms (nothing we can recover from)- >>> i haven't faced the issue of "host it too loaded" that status has >>> some other syndromes, however the behaviour on that state is very >>> much the same -waiting for 30 min (?) and than move it to activated. >>> Moran. >> >> 'host is too loaded' is too loaded is the only transient state where a >> temporary 'error' state makes sense, but in the same time, it can also >> fit the 'non operational' state description. >> From my experience, the problem with KVM/libvirt/VDSM mis-configured >> is never temporary, (= magically solved by itself, without concrete >> user intervention). IMHO, it should move the host to an error state >> that would not automatically recover from. >> Regardless, consolidating the names of the states ('inactive, >> detached, non operational, maintenance, error, unknown' ...) would be >> nice too. Probably can't be done for all, of course. >> Y. > > agreed, most of the causes of ERROR state aren't transient, but looks to > me as if this state is redundant and could be taken care as part of the > other host states, since the way it's being used today isn't very > helpful as well. > Moran. > Currently host status is changed to non-operational on various reasons, some of them are static like vdsm version and cpu model and some of them are (potentially) transient like network failure. The Error state, as Itamar detailed earlier on this thread, is used currently on what I would call (potentially) transient reason. The original intention (I think) was to move host to non-operational on reasons which are static and to Error on reasons which are transient, and I guess that is why there is timeout on the Error state and OE tries to initialize a host after 30 minutes in Error state. The problem is that as the code evolved this is not the case anymore. I suggest that we use the non-operational state for transient reasons, which we detect in monitoring flow, or execution failures and do the initialization retry as Laszlo suggested in the document. Use the Error state for static errors and remove the 'timeout' mechanism we currently have (from Error state).