storm99 Posted January 26, 2016 #1 Posted January 26, 2016 Hi, I have the following config: HP Microserver Gen8 16GB Ecc Ram HP P410 with battery and 512MB BBWC 4x WD 6TB RED Created a Raid-5 group within the HP P410, installed ESXi 6.0u1 (downloaded from HP), created a 16TB datastore using vSphere client 6. Tried fresh install of DSM 5.2 5592 with XPEnoboot_DS3615xs_5.2-5592.2, DSM 5.2 5644 with XPEnoboot_DS3615xs_5.2-5644.4 In both case, I created 4x2TB in the VM. DSM installed fine, did the upgrade, then create a Raid0/JBOD Disk Group / Volume without problem. Mounted share folders under windows and copied all my backups to those folders. But after using a day or two, the disk group /volume will crash - sometimes when copying files, sometime when unzipping files. Tried recreating the VM all over from scratch, and the crashing continues. Any idea? Thanks in advance!!!
brantje Posted January 26, 2016 #2 Posted January 26, 2016 Anything in the logs? (dmesg /var/log/syslog, regular services)
storm99 Posted January 26, 2016 Author #3 Posted January 26, 2016 By looking at the dmseg, I saw a lot of dm-1: WRITE SAME failed. Manually zeroing.
brantje Posted January 26, 2016 #4 Posted January 26, 2016 By searching on the error, the following interesting pages appeared: - https://bbs.archlinux.org/viewtopic.php?id=161843 - http://unix.stackexchange.com/questions ... e-esxi-5-5 Seems related to ESXi
brantje Posted January 26, 2016 #5 Posted January 26, 2016 Updated my above post, links were missing.
storm99 Posted January 26, 2016 Author #6 Posted January 26, 2016 thanks for the info. Fingers crossed After installing the open vm tools (open-vm-tools_bromolow-5.1_9.10.0-2476743-1.spk), the disk group /volume crash seems to dissappear. I am doing copy from Filestation, copying files to shared folder under Windows and unzipping files to shared folders, and no errors so far.
storm99 Posted January 26, 2016 Author #7 Posted January 26, 2016 Disk group / volume just crashed...
storm99 Posted January 27, 2016 Author #9 Posted January 27, 2016 Yes, physical disks are fine. XPenology use vmdk files as HDD.
brantje Posted January 27, 2016 #10 Posted January 27, 2016 Are those pre-allocated? Posted via MyXpenology
storm99 Posted January 28, 2016 Author #11 Posted January 28, 2016 Yes, tried both lazy zeroed and eager zeroed, same result
ferno Posted February 10, 2016 #12 Posted February 10, 2016 Hi, I have the exact same problem. Smaller size vmdk's seem to work fine though. This issue is driving me crazy. Can you post the exact setup of your vm? and which esx version you are running? Have you found a fix yet? The odd thing is that after a reboot the disk states as initialised and I can start rebuilding the volume but around 35% it crashes again.
brantje Posted February 12, 2016 #14 Posted February 12, 2016 Can you try raw device mapping? Posted via MyXpenology
storm99 Posted February 17, 2016 Author #15 Posted February 17, 2016 yes, pre-allocated. Lazy zeroed and eager zeroed makes no difference. DISK Group / Volume keep crashing
Tscherno Posted March 25, 2016 #16 Posted March 25, 2016 Hi, sorry for reactivating this old thread, but i have the exact same issue. I'm using xpenology in a VM under Windows. That setup ran smoothly for over a year, mainly using it as timemachine target. Every time i restarted Windows and forgot to shutdown DSM it gave me an unexpected shtudown, but never had a issue. Last week i got a crashed volume from that... OK, i thought my mistake should have shut it down before... Volume was only one Virt Disk. I took the oppertunity to upgrade to latest xpenology and DSM 5.2, and created a RAID 1 volume in hope that this will it make more stable. Took two days until volume crashed - without unexpected shutdown. Next try: RAID 5, again two days. Any ideas how to stabalize that? I learned pre-allocating the space will maybe help. Ok. But what about the maximum size? I'm using 2,4 TB volumes. Should i go for more but smaller?
ferno Posted March 26, 2016 #17 Posted March 26, 2016 Hi, Have been having this issue also, its driving me crazy as there is no consistency to it. At a certain point I thought I had figured it out and thought the culprit was filling up all 12 bays but after a while my new setup (with only 7 vmdk files) ended up with a crashed volume also.
b_lok Posted January 5, 2017 #18 Posted January 5, 2017 (edited) I got exactly the same problem as well! Does anyone know how to fix this issue? Thanks! Edited January 5, 2017 by Guest
Tscherno Posted January 5, 2017 #19 Posted January 5, 2017 I switched to Hyper-V instead of Vmware and never had an issue again!
Recommended Posts