[Spacewalk-list] Fix: Display system updates or packages failed

[Thomas von Steiger]

On Jul 25, 2008, at 5:50 AM, Justin Sherrill wrote:

> Thomas von Steiger wrote:
>> I get this errors if i try to display all available updates for a  
>> system.
>> Channels are centos5 and spacewalk is running on centos5 32bit.
>> yum update from the client site is running success for 15 available  
>> updates.
>> Are this maybe a timeout problem?
>> Can anybody view the available updates for a server?
>>
>> My virtual spacewalk server has 1Gb memory an 2 cpu's 3Ghz.
>> This are my oracle rpm's:
>>
>> oracle-instantclient-devel-10.2.0.4-1
>> oracle-xe-client-10.2.0.1-1.0
>> oracle-instantclient-basic-10.2.0.4-1
>> oracle-instantclient-jdbc-10.2.0.4-1
>> oracle-xe-univ-10.2.0.1-1.0
>> oracle-instantclient-sqlplus-10.2.0.4-1
>> oracle-lib-compat-10.2-3.fc9
>>
>> url from spacewalk:
>> https://spacewalk2.domain.local/rhn/systems/details/packages/UpgradableList.do?sid=1000010005
>>
>>
>> Errors in file /usr/lib/oracle/xe/app/oracle/admin/XE/udump/ 
>> xe_ora_4649.trc:
>> ORA-07445: exception encountered: core dump [evapls()+133]  
>> [SIGSEGV] [Address not mapped to object] [0x4] [] []
>>
>>
>> /usr/lib/oracle/xe/app/oracle/admin/XE/udump/xe_ora_4649.trc
>> Oracle Database 10g Express Edition Release 10.2.0.1.0 - Production
>> ORACLE_HOME = /usr/lib/oracle/xe/app/oracle/product/10.2.0/server
>> System name:    Linux
>> Node name:    spacewalk2.domain.local
>> Release:    2.6.18-92.1.6.el5xen
>> Version:    #1 SMP Wed Jun 25 14:54:06 EDT 2008
>> Machine:    i686
>> Instance name: XE
>> Redo thread mounted by this instance: 1
>> Oracle process number: 22
>> Unix process pid: 4649, image: oracleXE at spacewalk2.domain.local
>>
>> *** 2008-07-25 01:39:02.039
>> *** SERVICE NAME:(SYS$USERS) 2008-07-25 01:39:01.997
>> *** SESSION ID:(434.147) 2008-07-25 01:39:01.997
>> Exception signal: 11 (SIGSEGV), code: 1 (Address not mapped to  
>> object), addr: 0x4, PC: [0x9537687, evapls()+133]
>> Registers:
>> %eax: 0x00000004 %ebx: 0x280f8fd8 %ecx: 0x00000004
>> %edx: 0xb7f51a08 %edi: 0x00000006 %esi: 0x0c6df580
>> %esp: 0xbf87eacc %ebp: 0xbf87ec3c %eip: 0x09537687
>> %efl: 0x00210246
>>  evapls()+113 (0x9537673) jnz 0x9537ea0
>>  evapls()+119 (0x9537679) test $0x400,%eax
>>  evapls()+124 (0x953767e) jnz 0x9537e95
>>  evapls()+130 (0x9537684) mov 0xffffffe0(%ebp),%eax
>> > evapls()+133 (0x9537687) movzb (%eax),%edx
>>  evapls()+136 (0x953768a) testb $0x1,%dl
>>  evapls()+139 (0x953768d) jne 0x953769e
>>  evapls()+141 (0x953768f) push 0x8(%ebp)
>>  evapls()+144 (0x9537692) call 0xa142c2c
>> *** 2008-07-25 01:39:02.095
>> ksedmp: internal or fatal error
>> ORA-07445: exception encountered: core dump [evapls()+133]  
>> [SIGSEGV] [Address not mapped to object] [0x4] [] []
>> Current SQL statement for this session:
>> SELECT  pn.id AS id,
>>        pn.name AS name,
>>        pn.id || '|' || lookup_evr(full_list.evr.epoch,  
>> full_list.evr.version, full_list.evr.release) AS id_combo
>>  FROM  rhnPackageName PN,
>>        (
>>         SELECT  SOP.package_name_id AS name_id, MAX(PE.evr) evr
>>           FROM  rhnPackageEVR PE, rhnServerOutdatedPackages SOP
>>          WHERE  SOP.server_id = :1
>>            AND  SOP.package_evr_id = PE.id
>>       GROUP BY  SOP.package_name_id
>>         ) full_list
>> WHERE  full_list.name_id = PN.id
>> ORDER BY  UPPER(PN.name)
>> ----- Call Stack Trace -----
>> calling              call     entry                argument values  
>> in hex
>> location             type     point                (? means dubious  
>> value)
>> -------------------- -------- --------------------  
>> ----------------------------
>> ksedst()+27          call     ksedst1()            1 ? 1 ?
>> ksedmp()+557         call     ksedst()             1 ? 0 ? 0 ?  
>> 448F87 ?
>>                                                   FFFFFFFF ? 448F6C ?
>> ssexhd()+863         call     ksedmp()             3 ? 9537687 ?  
>> 70617665 ?
>>                                                   2928736C ?  
>> 3333312B ? 0 ?
>> evapls()+133         signal   00000000             B ? 44AC8C ?  
>> 44AD0C ?
>>
>> Or is this something for bugzilla?
>>
>> Thomas
>>
>>
>>
>>
>> _______________________________________________
>> Spacewalk-list mailing list
>> Spacewalk-list at redhat.com
>> https://www.redhat.com/mailman/listinfo/spacewalk-list
> Looks like you may be hitting an oracle bug.  See:
>
> https://bugzilla.redhat.com/show_bug.cgi?id=453176
>
> Looks like there may be a workaround as well..
>
> -Justin

Great!

SQL> alter system set "_optimizer_cost_based_transformation"=off  
scope=memory;
System altered.
SQL> commit;
Commit complete.

and the updates are there.

Thank you very much and have a nice day.

Thomas