[Spacewalk-list] Monitoring problem after 1.1 -> 1.2 update

[Benedetto Vassallo]

Def. Quota Benedetto Vassallo <vassallo at unipa.it>:

>
> Def. Quota Miroslav Suchý <msuchy at redhat.com>:
>
>> On 12/26/2010 05:59 PM, Benedetto Vassallo wrote:
>>> Hello list,
>>> I have a strange problem after upgrading my spacewalk server from 1.1 to
>>> 1.2
>>> In the monitoring page I see 2 spacewalk monitoring scouts, both having
>>> the same ssh key.
>>> One is "Fully Updated" (green triangle) and the other is "Failed" (red
>>> triangle).
>>>
>>> The result is all my monitored systems are "Awaiting updates" (clock icon).
>>>
>>> I searched in the oracle database for invalid objects (views, packaged
>>> or functions) but found nothing.
>>>
>>> How I can work around this?
>>
>> Hi,
>> first - this should not happened, I will be interested if you can  
>> reproduce it.
>>
> OK, I'll try to reproduce it on another spacewalk 1.1 installation  
> and then upgrading it to 1.2.
>
>
>> Fix has to be done directly in db.
>> Look in tables RHN_SAT_CLUSTER and RHN_SAT_NODE. There should be  
>> one record per scout, so you have there probably two records  
>> instead of one.
> I have 2 records in both tables.
>
> in RHN_SAT_CLUSTER the only difference is the 1st record is  
> associated to the "spaceadmin" user (the first user created after  
> install), the 2nd record is associated to "vassallo" (the user I use  
> normally in a different organization).
>
> SELECT RECID,TARGET_TYPE,CUSTOMER_ID,DESCRIPTION,LAST_UPDATE_USER,  
> physical_location_id, vip, deployed FROM RHN_SAT_CLUSTER;
>
> "RECID" "TARGET_TYPE"   "CUSTOMER_ID"   "DESCRIPTION"    
> "LAST_UPDATE_USER"      "PHYSICAL_LOCATION_ID"  "VIP"   "DEPLOYED"
> "1"     "cluster"       "1"     "Spacewalk Monitoring Scout"     
> "spaceadmin"    "1"     "147.163.2.24"  "1"
> "3"     "cluster"       "2"     "Spacewalk Monitoring Scout"     
> "vassallo"      "1"     "147.163.2.24"  "1"
>
> SELECT * FROM RHN_SAT_NODE;
>
> "RECID" "SERVER_ID"     "TARGET_TYPE"   "LAST_UPDATE_USER"       
> "LAST_UPDATE_DATE"      "MAC_ADDRESS"   "MAX_CONCURRENT_CHECKS"  
> "SAT_CLUSTER_ID"        "IP"    "SCHED_LOG_LEVEL"        
> "SPUT_LOG_LEVEL"        "DQ_LOG_LEVEL"  "SCOUT_SHARED_KEY"
> "2"     ""      "node"  ""      ""      "not set"       "10"    "1"   
>    "147.163.2.24"  "1"     "1"     "1"     "332709eb12bf"
> "4"     ""      "node"  ""      ""      "not set"       "10"    "3"   
>    "147.163.2.24"  "1"     "1"     "1"     "e854a34f7f77"
>
>
>
>> Delete one of them and then make sure that probes are asociate with  
>> correct scout (in webUI).
>>
>
> I deleted the one associated to "vassallo" becouse I see in my  
> production server (1.1) there is only the one assciated to the  
> "spaceadmin" user but it didn't work (red triangle after push).
>
> Tomorrow I try to reproduce it.
> The steps I did are:
> 1) Install spacewalk 1.1
> 2) Create the first user (spaceadmin)
> 3) Enable ldap authentication
> 4) Enable monitoring and monitoring scout
> 5) Create a new organization with a new user (vassallo) with  
> satellite admin rights
>
> ====== from now all operations are done by 'vassallo' user. ======
>
> 6) register some systems
> 7) Create a probe suite with a Linux -> CPU Usage probe and add all  
> systems to it
> 8) Update spacewalk to 1.2
>
> I'll let you know.
> Bye for now.
> -- 
> Benedetto Vassallo
> Sistema Informativo di Ateneo
> Settore Gestione Reti Hardware e Software
> U.O. Sviluppo e manutenzione dei sistemi
> Università degli studi di Palermo
>
> Red Hat Certified Engineer Certificate Number: 804007793324656
>
> Phone: +3909123860056
> Fax: +390916529124
>
> -------------------------------------------------------------------------
> This message was sent using the University of Palermo web mail interface.
>
>
> _______________________________________________
> Spacewalk-list mailing list
> Spacewalk-list at redhat.com
> https://www.redhat.com/mailman/listinfo/spacewalk-list
>

Hi Miroslav,
I probably understood what happened.
Today I have re-installed spacewalk 1.1 on a fresh CentOS 5.5 machine  
(x86_64) and executed all steps to reproduce the problem.
It happen after the upgrade process.
If I use the user 'vassallo' to enable monitoring and monitoring scout  
in web UI, then I have 2 scouts and nothing works.
If I use the user 'spaceadmin' (the first admin user created after the  
installation) to enable them, then I have 1 scout and all works fine.

I hope this can help you to fix it in future releases, or add a note  
in the upgrade guide.
Best regards
-- 
Benedetto Vassallo
Sistema Informativo di Ateneo
Settore Gestione Reti Hardware e Software
U.O. Sviluppo e manutenzione dei sistemi
Università degli studi di Palermo

Red Hat Certified Engineer Certificate Number: 804007793324656

Phone: +3909123860056
Fax: +390916529124

-------------------------------------------------------------------------
This message was sent using the University of Palermo web mail interface.