TOPIC:

Please help

Scenario: two brand new Dell T140 with PERC H330 hardware disk controllers, both servers with RAID1 Volumes created.

Issue: XCP-ng 7.6, clean installation and up to date packages (yum update)... during the DRBD installation process the local SR is not recognized... I've tried to pass the path using two different way but DRBD just cant take control of the volume!

Output:

---

| Convert local storage to iSCSI-HA storage? <yes or no> |

---

yes
Failed to locate SR for local storage

---

| Error converting local SR to usable block device |
| Enter full path to backing block device to be |
| used for HA storage and press <Enter> (ex. /dev/sdb) |

---

/dev/disk/by-id/scsi-364cd98f062544b00245993cf10381218-part3
Backing block device set to /dev/disk/by-id/scsi-364cd98f062544b00245993cf10381218-part3

---

| Enter LUN to be used for exposing iSCSI storage and press <Enter> |
| (Leave blank to accept default value of 10) |

---

LUN set to 10


Output (2nd) attempt, after redone clean XCP-ng + HAlizard installation process:

---

| Convert local storage to iSCSI-HA storage? <yes or no> |

---

yes
Failed to locate SR for local storage

---

| Error converting local SR to usable block device |
| Enter full path to backing block device to be |
| used for HA storage and press <Enter> (ex. /dev/sdb) |

---

/dev/mapper/XSLocalEXT--9db2df7a--0596--409f--a62d--7e1d27e7d46b-9db2df7a--0596--409f--a62d--7e1d27e7d46b
Backing block device set to /dev/mapper/XSLocalEXT--9db2df7a--0596--409f--a62d--7e1d27e7d46b-9db2df7a--0596--409f--a62d--7e1d27e7d46b

---

| Enter LUN to be used for exposing iSCSI storage and press <Enter> |
| (Leave blank to accept default value of 10) |

---

LUN set to 10

(...)

---

| Initializing DRBD.. This host will wait until its peer connects. |
| Installation will resume after the DRBD peers connect |

---

1+0 records in
1+0 records out
1048576 bytes (1,0 MB) copied, 0,00101544 s, 1,0 GB/s
open(/dev/mapper/XSLocalEXT--9db2df7a--0596--409f--a62d--7e1d27e7d46b-9db2df7a--0596--409f--a62d--7e1d27e7d46b) failed: Device or resource busy

Exclusive open failed. Do it anyways?
[need to type 'yes' to confirm] yes

open(/dev/mapper/XSLocalEXT--9db2df7a--0596--409f--a62d--7e1d27e7d46b-9db2df7a--0596--409f--a62d--7e1d27e7d46b) failed: Device or resource busy

Exclusive open failed. Do it anyways?
[need to type 'yes' to confirm] yes

md_offset 955076571136
al_offset 955076538368
bm_offset 955047391232

Found some data

==> This might destroy existing data! <==

Do you want to proceed?
[need to type 'yes' to confirm] yes

initializing activity log
initializing bitmap (28464 KB) to all zero
Writing meta data...
New drbd meta data block successfully created.
open(/dev/mapper/XSLocalEXT--9db2df7a--0596--409f--a62d--7e1d27e7d46b-9db2df7a--0596--409f--a62d--7e1d27e7d46b) failed: Device or resource busy

Exclusive open failed. Do it anyways?
[need to type 'yes' to confirm] yes

Starting DRBD resources: [
create res: iscsi1
prepare disk: iscsi1
adjust disk: iscsi1Operation canceled.
:failed(apply-al:20)
adjust net: iscsi1
]
..........

The are a couple of possible scenarios that could have led to the error
1)
The installer script can only handle conversion of a local SR when the local SR is backed by a single disk (or RAID being exposed as.a single disk). If you have selected more than one disk for local storage when installing xcp then the xcp installer would have created a volume group with multiple PVs backing it for the local SR. This situation poses a problem for the halizard installer because we are left with multiple block devices after deleting the local SR and the installer is written to handle only a single block device.

2)
We have seen rare cases where the halizard installer fails to delete the local SR that was created by the xcp installation. It's unlikely that this is the cause of your error.

In sum, if
1 above describes your situation, try reinstalling xcp and select only a single disk for local storage when prompted. Once the halizard installation is complete and the initial sync has finished, you can manually configure the second, third, etc disk to be managed by the halizard solution. There are several forum discussions on how to do this that I can send you if this is your scenario.

If (2) you only selected a single block device when installing xcp and encountered this error, you can get around it by manually deleting the local SR after installing xcp and then passing in the path to the backing block device when prompted by the halizard installer.

SOLVED!

Looks like Dell PERC H330 RAID controllers doesn't clear LVM2 metadata from the disks in the array, even destroying, recreating and reinitializing it from scratch.

I have to remove the disks and manually "wipefs --all --force /dev/sdX" on all disks out of the controller to solve this issue.

Thank you very much.