No relevant resource is found in the selected language.

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy>

Reminder

To have a better experience, please upgrade your IE browser.

upgrade

Setting all the public LUN as the snapshot reservation LUN causes to can’t apply any package

Publication Date:  2012-11-16 Views:  45 Downloads:  0
Issue Description
The cloud computing version: R2C00SPC200
After building the Galax environment, apply a virtual machine package through the cloud computing user’s self-service system, but the package is soon returned from the cloud computing operations support system: the verification passes, but assignment fails.
The detail information is:
R201203300001 Win2003Taocan1 1.00 1 2012-03-30 09:48:02 test verification passes,the assignment fails  admin 2012-03-30 09:48:24

Alarm Information
None
Handling Process
1. Execute “bms status” on the ESC to check the status and it displays “normal”.
2. There are two nodes at site, apply package again, just apply one virtual machine, we meet the same problem.
3. Analyze the logs of the ESC, from the “” log, we can see when we creating one virtual machine, the package application is successful, but the assignment is failed.
<182> 6 2012-03-30T09:48:31.448769+00:00   {INFO } {com.huawei.galax.bms.listener.OrderManagerListener 646} R201203300001 create vmcount is :1
<182> 6 2012-03-30T09:48:31.448936+00:00   {INFO } {com.huawei.galax.bms.listener.OrderManagerListener 652} vm's alias :win2003taocan2.
<182> 6 2012-03-30T09:48:31.666801+00:00   {INFO } {com.huawei.galax.bms.listener.OrderManagerListener 1425} successed in assigning resource: order no.R201203300001
<180> 4 2012-03-30T09:49:31.592413+00:00   {WARN } {com.huawei.galax.bms.listener.OrderManagerListener 1483} creating resource is not successed: order no.R201203300001
4. Analyze the “cloud-debug.log” furthermore, we find the cause is “There is no public lun in DB,please check it”, it’s caused by there isn’t public lun.
2012 Mar 30 09:48:31,653 {INFO } {ClusterSink.6} {edu.ucsb.eucalyptus.cloud.ws.VolumeSynSender 81} VolumeManager CreateVolumeSystem begin
2012 Mar 30 09:48:31,653 {DEBUG} {ClusterSink.6} {edu.ucsb.eucalyptus.cloud.ws.VolumeSynSender 95} volume sum :1
2012 Mar 30 09:48:31,654 {DEBUG} {ClusterSink.6} {edu.ucsb.eucalyptus.cloud.ws.VolumeSynSender 96} generate volume size :80
2012 Mar 30 09:48:31,654 {INFO } {ClusterSink.6} {edu.ucsb.eucalyptus.cloud.ws.VolumeManager 3081} Method into createVolumeId, launchIndex: 4
2012 Mar 30 09:48:31,657 {DEBUG} {ClusterSink.6} {edu.ucsb.eucalyptus.cloud.ws.VolumeSynSender 102} volume Count  :0
2012 Mar 30 09:48:31,657 {DEBUG} {ClusterSink.6} {edu.ucsb.eucalyptus.cloud.ws.VolumeSynSender 103} generate volume Id :vol-286E063E
2012 Mar 30 09:48:31,658 {DEBUG} {ClusterSink.6} {edu.ucsb.eucalyptus.cloud.ws.VolumeSynSender 407} VolumeSynSender CreateSingleVolume begin to search ClusterInfo_clusterInfoId!
2012 Mar 30 09:48:31,659 {DEBUG} {ClusterSink.6} {com.huawei.galax.storage.vbs.client.BRMClient 119} BRMClient createVolume Enter
2012 Mar 30 09:48:31,659 {DEBUG} {ClusterSink.6} {com.huawei.galax.storage.vbs.client.BRMClient 120} CreateStorageVolumeType request information: volumeID=vol-286E063E,size=80,snapshotID=null, StorageType=1, getStoragePosition=null
2012 Mar 30 09:48:31,659 {DEBUG} {ClusterSink.6} {com.huawei.galax.storage.vbs.client.BRMClient 137} BRMClient createVolume service targetEndpoint:http://192.168.8.33:8550/brm/services/BrmServices
2012 Mar 30 09:48:31,704 {ERROR} {ClusterSink.6} {com.huawei.galax.storage.vbs.client.BRMClient 153} BRM createVolume failed! --AxisFault
org.apache.axis2.AxisFault: There is no public lun in DB,please check it.

5. Known from the site, after building the Galax according to the GPI document, they dispensed the configuration, and then configured the snapshot reservation room according to the initialization configuration in the GPI document, set the two LUNs both as private LUN.
6. Modify the snapshot LUN as public LUN, and then restart the BMS process, create the package successfully. As the following example:
a) Execute: “pg_dump -D -U postgres vbsdb >/home/VBSdb_Bak_2011_10_10.sql”, after it has been executed completely, check whether the spare database document has produced normally in the “/home” catalog;
b) Login in the master CRM database via the “pgAdmin 3” tool, check the “luntab” table, find the value of “attribute” is “snapshot”, then record its lun_id;
Modify the snapshot LUN’s property as to public LUN via SQL sentence:
update luntab set attribute = 'public' where lun_id = “lun_id”
c) Restart the BRM process of the CRM master node
Login in the CRM master node via putty, execute the command: brm restart.
Root Cause
1. The BMS Portal process’s problem.
2. The number of the virtual machines contained in the application package has exceeded the number of the CNA nodes.
3. There isn’t available resource.

Suggestions
There is detail description of the LUN’s function in the GPI manual:
“When taking the IP SAN storage, the system default doesn’t assign the LUN in the IP SAN to snapshot.
Use the snapshot function: set the number of storage snapshot’s LUNs.
Not use the snapshot function: don’t need to set the number, the reserved number of the LUN to deposit the snapshot is default as 0. The LUN reserved for the snapshot is just used as the snapshot’s storage room, it won’t assign storage resource for other functions.”
We need to understand the corresponding content of the GPI carefully when we configuring at the site.

END