It seems that CS 4.6 does not have enough logic to handle correctly the snapshots. In my case, it was an expunged volume that used to create the problem as snapshots of this volume still existed (both in DB and in secondary storage). So I created a manual workaround to bypass this problem by making our hands dirty in MySQL:
- Stop CS service - In MySQL find the expunged volume in “volumes" table and keep the volume_id - In “snapshot_store_ref" table get the location of the snapshots by using the mentioned volume_id. Then go to the secondary storage and delete all the relevant files (since now we know the exact location) - Now it’s time to update the tables by running the following SQL commands: <update snapshot_store_ref set state="Destroyed" where id=XXX;> <update snapshots set status="Destroyed" where id=XXX;> where XXX is the volume id of the second step. - Start CS service and check the snapshots output. The problem should be gone. Hope that helps some fellows. Regards Stavros ---------------------------- Stavros Konstantaras Science faculty Research IT support (FEIOG) University of Amsterdam, Science Park 904, 1098 XH Fingerprint: E5E5 9B19 D1CD 88CD 4763 3465 A8DC 7C92 330F D59A > On 16 Mar 2016, at 11:05, Stavros Konstantaras <s.konstanta...@uva.nl> wrote: > > Hi Adrian, > > Firstly, thanks for your extensive answer and the trick. My problem is > similar to the CLOUDSTACK-8845 issue. In our case we just try to view the > snapshots from cloudstack UI instead of listing them in cloud monkey. My > first thoughts are that probably some inconsistencies exist in the database > but unfortunately after running the following mysql commands I don’t get any > results. > > < select volume_id,name,status from cloud.snapshots where volume_id not in > (select id from cloud.volumes); > > < select * from cloud.snapshot_store_ref where volume_id not in (select id > from cloud.volumes); > > > Any other ideas to find out the inconsistencies and get rid of them? In > addition, I checked the log file of cloudstack-management server and I don’t > see the error you mentioned. I use CS 4.6 and XenServer 6.5 > > Thanks again. > > Regards > Stavros > > ---------------------------- > Stavros Konstantaras > Science faculty Research IT support (FEIOG) > University of Amsterdam, Science Park 904, 1098 XH > > Fingerprint: E5E5 9B19 D1CD 88CD 4763 3465 A8DC 7C92 330F D59A > >> On 16 Mar 2016, at 07:43, Adrian Sender <asen...@testlabs.com.au> wrote: >> >> Hi Stravos, >> >> A bit more info with ongoing snapshot issues with Xenserver.. your snapshots >> may also be failing if your chain depth exceeds 30. >> >> Assuming your using LVMoHBA (hardware hba), easiest way to check this is as >> follows: >> >> If the xenserver snapshot chain depth exceeds 30, you will not be able to >> take >> any new snapshots. >> >> You may see a similar error: (note exact error will depend on CCP / CS >> version.. >> >> 2015-11-18 10:57:42,133 WARN [c.c.h.x.r.XenServerStorageProcessor] >> (DirectAgent-243:ctx-341ac146) (logid:65b51cbd) create snapshot operation >> Failed for snapshotId: 36709, reason: SR_BACKEND_FAILURE_109The snapshot >> chain >> is too long >> SR_BACKEND_FAILURE_109The snapshot chain is too long >> >> >> Example >> ======= >> [root@qh2-nsp04 ~]# vhd-util query -vsfd -n >> /dev/VG_XenStorage-3965cb14-960e-d9d5-4ab9-608425f3ee20/VHD-7bf051ee-7762-4902-84be-4962e70a4e3e >> -p >> 10240 >> 1145303552 >> /dev/mapper/VG_XenStorage--3965cb14--960e--d9d5--4ab9--608425f3ee20-VHD--811a34ff--731a--4c36--92be--286831ab7393 >> hidden: 0 >> chain depth: 30 >> >> The best way to get the LVM path is as follows: >> >> mysql> select * from volumes where name='ROOT-3230' \G >> >> We are looking for the path, in this case >> "7bf051ee-7762-4902-84be-4962e70a4e3e" >> >> Login to Xenserver console and do "lvscan | grep >> 7bf051ee-7762-4902-84be-4962e70a4e3e" This will give us VG and LV. >> >> Then you can execute vhd-util query -vsfd -n >> /dev/VG_XenStorage-3965cb14-960e-d9d5-4ab9-608425f3ee20/VHD-e00cd7a7-ae20-4b5d-a0e3-1c215d432ed4 >> >> Delete the broken snapshots from the primary storage in XenCenter. >> >> Regards, >> Adrian Sender >> >> Testlabs Australia >> Research Platform Services >> M: +61. 487-440-732 >> E: asen...@testlabs.com.au >> W: http://www.testlabs.com.au >> >> ---------- Original Message ----------- >> From: "Adrian Sender" <asen...@testlabs.com.au> >> To: users@cloudstack.apache.org >> Sent: Wed, 16 Mar 2016 09:59:34 +1000 >> Subject: Re: [Where are the snapshot file] >> >>> Hi Stravos, >>> >>> Jumping in a bit late on this thread...Can you share the exact >>> management log error. >>> >>> We have ran CCP 4.3 with Xenserver, everything is generally ok; most >>> common snapshot issues are as follows: >>> >>> Snapshot failing >>> ---------------- >>> >>> The snapshotting on data volume DATA-3230 Failing. UI shows "Error" >>> >>> 2014-08-26 10:51:32,884 DEBUG [o.a.c.s.m.AncientDataMotionStrategy] >>> >>> (Work-Job-Executor-6:ctx-274b0a29 job-31179/job-31180 ctx-8511b016) >>> copyAsync >>> inspecting src type SNAPSHOT copyAsync inspecting dest type SNAPSHOT >>> 2014-08-26 10:51:33,055 DEBUG [o.a.c.s.m.AncientDataMotionStrategy] >>> >>> (Work-Job-Executor-6:ctx-274b0a29 job-31179/job-31180 ctx-8511b016) copy >>> snasphot failed: java.lang.NullPointerException >>> 2014-08-26 10:51:33,056 DEBUG [o.a.c.s.m.AncientDataMotionStrategy] >>> >>> (Work-Job-Executor-6:ctx-274b0a29 job-31179/job-31180 ctx-8511b016) >>> copy failed com.cloud.utils.exception.CloudRuntimeException: >> java.lang.NullPointerException >>> >>> 1. Show the details of the failing snapshot volume >>> >>> mysql> select * from volumes where name=DATA-3230 \G >>> >>> 2. get the id of the volume from the above query and delete >>> snapshots in the 'Ready' state. >>> >>> delete from snapshot_store_ref where volume_id=3934 and state='Ready' >>> >>> 3. Snapshots will now succeed for volume DATA-3230. >>> >>> Snapshot Failing >>> ---------------- >>> >>> You may see the following in the logs: >>> >>> 2015-11-17 15:22:16,586 DEBUG [c.c.h.x.r.XenServerStorageProcessor] >>> (DirectAgent-339:ctx-6465ce91) (logid:9224bc6d) Failed to destroy pbd >>> SR_BACKEND_FAILURE_40The SR scan failed [opterr=['INTERNAL_ERROR', >>> 'Db_exn.Uniqueness_constraint_violation("VDI", "uuid", >>> "e5318c7f-9434-4f1c-8695-3e82a69d2192")']] >>> >>> http://support.citrix.com/article/CTX135230 >>> >>> Note in this example DATA-3701 snapshot is failing, but the >>> screenshot shows the NFS secondary storage mount for account 29, >>> DATA-3701 belongs to account 113. So you can go through the tabs >>> until you find /var/cloud_mount/xxxx/snapshots/113/3701 and unplug >>> just that volume. >>> >>> In this example you will see many attached NFS secondary storage >>> volumes on the physical xenserver host. >>> >>> 1. Get the uuid of the SR from the "General" tab on the NFS storage mount. >>> >>> xe sr-list uuid=84142bd7-c19c-4a39-850b-ddd30539d8ed >>> >>> [root@qh2-nsp01 ~]# xe sr-list uuid=84142bd7-c19c-4a39-850b-ddd30539d8ed >>> uuid ( RO) : 84142bd7-c19c-4a39-850b-ddd30539d8ed >>> name-label ( RW): >>> 14a68eb4-62aa-4beb-91d5-e5070ecd46a9/var/cloud_mount/b9d64979-0c47- >>> 31fe-a496-6943e9dfca63/snapshots/29/4469 name-description ( RW): >>> 14a68eb4-62aa-4beb-91d5-e5070ecd46a9/var/cloud_mount/b9d64979-0c47- >>> 31fe-a496-6943e9dfca63/snapshots/29/4469 host ( RO): >>> qh2-nsp01.nsp.nectar.org.au type ( RO): file >>> content-type ( RO): file >>> >>> 2. Get the PBD UUID by using the sr-uuid from above >>> >>> [root@qh2-nsp01 ~]# xe pbd-list sr-uuid=84142bd7-c19c-4a39-850b-ddd30539d8ed >>> uuid ( RO) : c15627c7-a968-ac1f-9794-2e3d66c0cec4 >>> host-uuid ( RO): 14a68eb4-62aa-4beb-91d5-e5070ecd46a9 >>> sr-uuid ( RO): 84142bd7-c19c-4a39-850b-ddd30539d8ed >>> device-config (MRO): location: >>> /var/cloud_mount/b9d64979-0c47-31fe-a496-6943e9dfca63/snapshots/29/4469 >>> currently-attached ( RO): true >>> >>> 3. Unplug the pdb using the uuid from above >>> >>> xe pbd-unplug uuid=c15627c7-a968-ac1f-9794-2e3d66c0cec4 >>> >>> 4. At this point you should notice the state change to broken and a >>> red cross through the secondary mount point in Xencenter. >>> >>> 5. Finally remove the sr from the host completely using the original >>> sr uuid. >>> >>> xe sr-forget uuid=84142bd7-c19c-4a39-850b-ddd30539d8ed >>> >>> 6 Repeat this procedure for any other attached NFS snapshot mounts, >>> check all xenservers. >>> >>> Regards, >>> Adrian Sender >>> >>> ---------- Original Message ----------- >>> From: Prashant Mishra <prashant.mis...@accelerite.com> >>> To: "users@cloudstack.apache.org" <users@cloudstack.apache.org> >>> Sent: Tue, 15 Mar 2016 12:48:10 +0000 >>> Subject: Re: [Where are the snapshot file] >>> >>>> There are some discussion related to this please check >>>> https://issues.apache.org/jira/browse/CLOUDSTACK-9297 >>>> >>>> On 3/15/16, 5:49 PM, "Stavros Konstantaras" <s.konstanta...@uva.nl> wrote: >>>> >>>>> Yes indeed. Currently I am fighting with the very well known "Unable to >>>>> determine the storage pool of the snapshot² after deleting some snapshots >>>>> and creating new ones. >>>>> >>>>> Is there any workaround for this issue? >>>>> >>>>> Cheers >>>>> Stavros >>>>> >>>>> ---------------------------- >>>>> Stavros Konstantaras >>>>> Science faculty Research IT support (FEIOG) >>>>> University of Amsterdam, Science Park 904, 1098 XH >>>>> >>>>> Fingerprint: E5E5 9B19 D1CD 88CD 4763 3465 A8DC 7C92 330F D59A >>>>> >>>>>> On 15 Mar 2016, at 13:01, Simon Weller <swel...@ena.com> wrote: >>>>>> >>>>>> Can you post some logs from both the management server and the agent >>>>>> (on the host) when you try and delete a snapshot? >>>>>> >>>>>> I just looked through Jira and there are quite a few issues related to >>>>>> snapshots. >>>>>> >>>>>> - Si >>>>>> >>>>>> ________________________________________ >>>>>> From: Gian Paolo Buono <gianpaolo.bu...@gesca.it> >>>>>> Sent: Tuesday, March 15, 2016 6:52 AM >>>>>> To: users@cloudstack.apache.org >>>>>> Subject: Re: [Where are the snapshot file] >>>>>> >>>>>> Hi Simon, >>>>>> >>>>>> we run Citrix CloudPlatfor 4.3 but we want to migrate to CloudStack >>>>>> 4.8... >>>>>> >>>>>> there is a fix ? >>>>>> >>>>>> Best regards >>>>>> >>>>>> >>>>>> >>>>>> This communication may contain information that is proprietary, >>>>>> confidential, or exempt from disclosure. If you are not the intended >>>>>> recipient, please note that any other dissemination, distribution, use >>>>>> or copying of this communication is strictly prohibited. Anyone who >>>>>> receives this message in error should notify the sender immediately by >>>>>> telephone or by return e-mail and delete it from his or her computer. >>>>>> >>>>>> On 03/15/2016 12:46 PM, Simon Weller wrote: >>>>>> >>>>>> Gian, >>>>>> >>>>>> What version of CloudStack are you running? >>>>>> >>>>>> I recall (quite a long time ago), there was a bug where volumes and >>>>>> snapshots weren't getting cleaned up correctly. >>>>>> If I recall correctly, that was somewhere around 4.2. >>>>>> >>>>>> - Si >>>>>> >>>>>> ________________________________________ >>>>>> From: Gian Paolo Buono >>>>>> <gianpaolo.bu...@gesca.it><mailto:gianpaolo.bu...@gesca.it> >>>>>> Sent: Tuesday, March 15, 2016 6:10 AM >>>>>> To: users@cloudstack.apache.org<mailto:users@cloudstack.apache.org> >>>>>> Subject: Re: [Where are the snapshot file] >>>>>> >>>>>> Thank you it's correct :) , but now I have another question.. >>>>>> >>>>>> From console of CloudStack I have deleted a snapshot, on the database >>>>>> the status is changed in Destroyed but the file it's stil on secondary >>>>>> storage: >>>>>> >>>>>> ls -ltr d8736bdb-22c6-46fe-a27e-0ee5dc99a867.vhd >>>>>> -rw-r--r-- 1 root root 1271296512 Mar 11 04:02 >>>>>> d8736bdb-22c6-46fe-a27e-0ee5dc99a867.vhd >>>>>> >>>>>> Also in snapshots view sometimes the status is allocated...but what >>>>>> does it mean ? >>>>>> >>>>>> >>>>>> Best Regards >>>>>> >>>>>> On 03/15/2016 11:15 AM, Stavros Konstantaras wrote: >>>>>> >>>>>> Hi Gian, >>>>>> >>>>>> If you run the following query in MySQL and check the column >>>>>> ³install_path² you will find the location of snapshot file in secondary >>>>>> storage: >>>>>> <select * from cloud.snapshot_store_ref;> >>>>>> >>>>>> I hope that it helps you. >>>>>> >>>>>> Best Regards >>>>>> Stavros >>>>>> >>>>>> >>>>>> ---------------------------- >>>>>> Stavros Konstantaras >>>>>> Science faculty Research IT support (FEIOG) >>>>>> University of Amsterdam, Science Park 904, 1098 XH >>>>>> >>>>>> Fingerprint: E5E5 9B19 D1CD 88CD 4763 3465 A8DC 7C92 330F D59A >>>>>> >>>>>> >>>>>> >>>>>> On 15 Mar 2016, at 11:11, Gian Paolo Buono >>>>>> <gianpaolo.bu...@gesca.it><mailto:gianpaolo.bu...@gesca.it><mailto:gianpa >>>>>> olo.bu...@gesca.it><mailto:gianpaolo.bu...@gesca.it> wrote: >>>>>> >>>>>> Hi all, how can I know where are the snapshot file (vhd) on the >>>>>> secondary storage ? Is there a mysql query ? >>>>>> >>>>>> Kind regards >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>> >>>> >>>> DISCLAIMER >>>> ========== >>>> This e-mail may contain privileged and confidential information >>>> which is the property of Accelerite, a Persistent Systems business. >>>> It is intended only for the use of the individual or entity to which >>>> it is addressed. If you are not the intended recipient, you are not >>>> authorized to read, retain, copy, print, distribute or use this >>>> message. If you have received this communication in error, please >>>> notify the sender and delete all copies of this message. Accelerite, >>>> a Persistent Systems business does not accept any liability for >>>> virus infected mails. >>> ------- End of Original Message ------- >> ------- End of Original Message ------- >> >
