Skip to main content
Start a Conversation

This post is more than 5 years old

Solved!

Go to Solution

AGerasimov

8 Posts

7457

August 3rd, 2017 11:00

Add SDS device IO error

Hello.

I have a strange error from SDS. I was trying ScaleIO 2.0.1.3 SDS on XenServer7 host. At first I have successfully added 240GB SSD to SDS and it was working fine, Then I have noticed that the disk was presented as RAID0 single disk, not as JBOD, so I removed it from SDS, changed it to JBOD (without rebooting the host) and was trying to introduce it again to ScaleIO, but received "SDS device IO error". I have tried to write/read to the disk directly from the host and had no errors at all. I restarted SDS - nothing, rebooted host - nothing. The disk is perfectly accessible from dom0. What could cause this error?

RHasleton1

73 Posts

August 11th, 2017 08:00

Hi AGerasimov,

The errors you highlighted above indicate that the SDS is having issues writing to the disk while trying to initialize and add the disk. This is a new requirement/feature added in 2.0.1.3 code to make sure that slow disks are not added to the cluster. You can see the details here in the 2.0.1.3 release notes. Search for SCI-21795.

There is a time-out of 4.5 seconds given to the new disk to pass muster, so to speak. The beginning timestamp is 13:00:12.983359 and the ending where it times out is 13:00:17.446133, which is about 4.5 seconds. You can change the timeout value as stated in the release notes, but I would trust the error at this point.

If it worked when in RAID0 mode, but not in JBOD mode, then check with the card vendor to verify that something is not wrong with the config there.

Hope that helps

pawelw1

306 Posts

August 4th, 2017 02:00

Hi,

Try to use "--force_device_takeover" switch when trying to add an SDS device, or try to wipe it with dd (dd if=/dev/zero of=/dev/disk) - at least first few GBs - and see if you can add it then.

Also, we would rather recommend the opposite - every SDS device should be standalone RAID0 instead of JBOD, so you might want to change it again ;-)

Cheers,

Pawel

AGerasimov

8 Posts

August 4th, 2017 07:00

Pawel, I have tried to force device takeover and also have zeroed the whole disk with dd without any luck.

Why do you recommend RAID0? From LSI point of view they recommend JBOD for better latency. According to their test results JBOD could provide almost twice IOPS compared to RAID0.

I downloaded logs from SDS. Maybe this could help a bit

3/08 13:00:12.981311 0x7f43c8ea2eb0:contCmd_AddDevGroup:03243: Add Dev group start bReconfigInProgress 0, bUpdateDevicePath 1 # of devices 1

03/08 13:00:12.981327 0x7f43c8ea2eb0:mos_StartBlkDevEnum:00795: open partition file: /proc/partitions

03/08 13:00:12.981449 0x7f43c8ea2eb0:mos_StartBlkDevEnum:00795: open partition file: /proc/partitions

03/08 13:00:12.981873 0x7f43c9c4feb0:mosAsyncIO_OpenFileEx:00376: WARNING: Failed to open IO file /dev/sda1 with rc 35

03/08 13:00:12.981882 0x7f43c9c4feb0:file_OpenEx:00694:  Open error /dev/sda1, DEV_MOUNTED

03/08 13:00:12.981885 0x7f43c9c4feb0:phyDev_ReadDevId:02647: failed to open file: path=/dev/sda1, DEV_MOUNTED

03/08 13:00:12.981889 0x7f43c9c4feb0:phyDevMap_ThreadedReadDevId:00736: Failed to read DeviceId of /dev/sda1. rc=35 took 0 milli

03/08 13:00:12.981894 0x7f43c9c4feb0:contCmd_AddDevGroupFilter:03189: Compare new dev /dev/sdb with /dev/sda1

03/08 13:00:12.982232 0x7f43c9c46eb0:mosAsyncIO_CloseFileIntern:00503: Closing device path:/dev/sda2

03/08 13:00:12.982242 0x7f43c9c46eb0:mosIO_close:00115: Closing FD(26)

03/08 13:00:12.982284 0x7f43c9c46eb0:phyDev_ReadDevId:02679: Invalid device header signature : path=/dev/sda2, devVersion=0, sigStart=0, sigEnd=0

03/08 13:00:12.982301 0x7f43c9c46eb0:phyDevMap_ThreadedReadDevId:00736: Failed to read DeviceId of /dev/sda2. rc=351 took 0 milli

03/08 13:00:12.982305 0x7f43c9c61eb0:mosAsyncIO_CloseFileIntern:00503: Closing device path:/dev/sda5

03/08 13:00:12.982326 0x7f43c9c61eb0:mosIO_close:00115: Closing FD(30)

03/08 13:00:12.982335 0x7f43c9c46eb0:contCmd_AddDevGroupFilter:03189: Compare new dev /dev/sdb with /dev/sda2

03/08 13:00:12.982354 0x7f43c9c61eb0:phyDev_ReadDevId:02679: Invalid device header signature : path=/dev/sda5, devVersion=3800800028008, sigStart=1800800008008, sigEnd=0

03/08 13:00:12.982362 0x7f43c9c7ceb0:mosAsyncIO_CloseFileIntern:00503: Closing device path:/dev/loop0

03/08 13:00:12.982380 0x7f43c9c61eb0:phyDevMap_ThreadedReadDevId:00736: Failed to read DeviceId of /dev/sda5. rc=351 took 0 milli

03/08 13:00:12.982386 0x7f43c9c73eb0:mosAsyncIO_CloseFileIntern:00503: Closing device path:/dev/sda

03/08 13:00:12.982394 0x7f43c9c73eb0:mosIO_close:00115: Closing FD(27)

03/08 13:00:12.982398 0x7f43c9c61eb0:contCmd_AddDevGroupFilter:03189: Compare new dev /dev/sdb with /dev/sda5

03/08 13:00:12.982410 0x7f43c9c73eb0:contCmd_AddDevGroupFilter:03189: Compare new dev /dev/sdb with /dev/sda

03/08 13:00:12.982421 0x7f43c9c7ceb0:mosIO_close:00115: Closing FD(24)

03/08 13:00:12.982434 0x7f43c9c7ceb0:phyDev_ReadDevId:02679: Invalid device header signature : path=/dev/loop0, devVersion=20202032336e6957, sigStart=1313030444301, sigEnd=0

03/08 13:00:12.982440 0x7f43c9c7ceb0:phyDevMap_ThreadedReadDevId:00736: Failed to read DeviceId of /dev/loop0. rc=351 took 0 milli

03/08 13:00:12.982443 0x7f43c9c7ceb0:contCmd_AddDevGroupFilter:03189: Compare new dev /dev/sdb with /dev/loop0

03/08 13:00:12.982554 0x7f43c9c85eb0:mosAsyncIO_CloseFileIntern:00503: Closing device path:/dev/sdb

03/08 13:00:12.982564 0x7f43c9c85eb0:mosIO_close:00115: Closing FD(26)

03/08 13:00:12.982581 0x7f43c9c85eb0:phyDev_ReadDevId:02679: Invalid device header signature : path=/dev/sdb, devVersion=0, sigStart=0, sigEnd=0

03/08 13:00:12.982588 0x7f43c9c85eb0:phyDevMap_ThreadedReadDevId:00736: Failed to read DeviceId of /dev/sdb. rc=351 took 0 milli

03/08 13:00:12.982592 0x7f43c9c85eb0:contCmd_AddDevGroupFilter:03189: Compare new dev /dev/sdb with /dev/sdb

03/08 13:00:12.982610 0x7f43c8ea2eb0:phyDevMap_ReloadSpecificUpdate:00227: Found device /dev/sda ,ebd5818e00040000

03/08 13:00:12.982632 0x7f43c9c58eb0:mosAsyncIO_CloseFileIntern:00503: Closing device path:/dev/sda3

03/08 13:00:12.982650 0x7f43c9c58eb0:mosIO_close:00115: Closing FD(29)

03/08 13:00:12.982658 0x7f43c9c58eb0:phyDev_ReadDevId:02679: Invalid device header signature : path=/dev/sda3, devVersion=0, sigStart=0, sigEnd=0

03/08 13:00:12.982667 0x7f43c9c58eb0:phyDevMap_ThreadedReadDevId:00736: Failed to read DeviceId of /dev/sda3. rc=351 took 0 milli

03/08 13:00:12.982672 0x7f43c8ea2eb0:contCmd_AddDevGroup:03325: Created 0 UMT's for Non-New devices - waiting for their completion

03/08 13:00:12.982679 0x7f43c8ea2eb0:contCmd_AddDevGroup:03331: Creating UMT's for New devices

03/08 13:00:12.982683 0x7f43c8ea2eb0:contCmd_AddDevGroup:03355: Created 1 UMT's for New devices - waiting for their completion

03/08 13:00:12.982688 0x7f43c9c58eb0:contCmd_AddDev:01214: DevId ffdff6ca00010000 - Start rc = SUCCESS

03/08 13:00:12.982692 0x7f43c9c58eb0:contCmd_AddDev:01277: Throttling info: rebuildConcurrentIOLimit=1, rebalanceConcurrentIOLimit=1, rebuildBWLimitKB=0, rebalanceBWLimitKB=10240, rebuildQuietPeriodMs=0, rebalanceQuietPeriodMs=0, rebuildAppIopsPerDeviceThreshold=0, rebalanceAppIopsPerDeviceThreshold=0, rebuildAppBwPerDeviceThresholdKB=0, rebalanceAppBwPerDeviceThresholdKB=0

03/08 13:00:12.982733 0x7f43c9c85eb0:contDev_BackgroundScan:00334: Starting contDev_BackgroundScan for device 0x7f43b8004960

03/08 13:00:12.983096 0x7f43c9c7ceb0:mosAsyncIO_CloseFileIntern:00503: Closing device path:/dev/sdb

03/08 13:00:12.983105 0x7f43c9c7ceb0:mosIO_close:00115: Closing FD(18)

03/08 13:00:12.983132 0x7f43c9c7ceb0:phyDev_ReadDevId:02679: Invalid device header signature : path=/dev/sdb, devVersion=0, sigStart=0, sigEnd=0

03/08 13:00:12.983138 0x7f43c9c7ceb0:phyDevMap_ThreadedReadDevId:00736: Failed to read DeviceId of /dev/sdb. rc=351 took 10 milli

03/08 13:00:12.983342 0x7f43c9c58eb0:AddDevProgress_Trace:00255: AddDevProgress: osThreadId=4af, devId=ffdff6ca00010000, Start=aa2aabf7a [usec], end=aa2aac03c [usec], interval=     194 [usec], Desc=Device Open

03/08 13:00:12.983351 0x7f43c9c58eb0:contDev_AddIntern:00800: Device /dev/sdb sector size: 512

03/08 13:00:12.983356 0x7f43c9c58eb0:phyDev_InitWithBuf:01976: Verifying device size: 1bf24000 to min/max: b400000/440000000

03/08 13:00:12.983359 0x7f43c9c58eb0:phyDev_InitWithBuf:01997: DevInit: sizeInLbs=1bf24000, dirtyGranularity=2048, tgtId=9a73379b00000001, devId=ffdff6ca00010000, bDevZeroPadEnabled=0, bChecksumEnabled=0

03/08 13:00:12.983362 0x7f43c9c58eb0:mos_StartBlkDevEnum:00795: open partition file: /proc/partitions

03/08 13:00:12.983509 0x7f43c9c58eb0:phyDev_PrintOffsets:01470: Offsets: 96 - 8192 - 409600 - 409712 - 409824 - 2097152

03/08 13:00:12.983515 0x7f43c9c58eb0:phyDev_PrintOffsets:01480: Sizes: devSize 468860928 number of combs 5120 number of teeth 227911 dirty granularity 2048,BM mode memory

03/08 13:00:12.985451 0x7f43c9c3deb0:mosAsyncIO_CloseFileIntern:00503: Closing device path:/dev/tda

03/08 13:00:12.985465 0x7f43c9c3deb0:mosIO_close:00115: Closing FD(24)

03/08 13:00:12.985494 0x7f43c9c3deb0:phyDev_ReadDevId:02679: Invalid device header signature : path=/dev/tda, devVersion=7ccbf54d0a15447, sigStart=7b625f65e97a7e2a, sigEnd=d4c013a4a3cf3aa5

03/08 13:00:12.985504 0x7f43c9c3deb0:phyDevMap_ThreadedReadDevId:00736: Failed to read DeviceId of /dev/tda. rc=351 took 10 milli

03/08 13:00:13.039449 0x7f43c9c8eeb0:mosAsyncIO_CloseFileIntern:00503: Closing device path:/dev/sda4

03/08 13:00:13.039459 0x7f43c9c8eeb0:mosIO_close:00115: Closing FD(25)

03/08 13:00:13.039489 0x7f43c9c8eeb0:phyDev_ReadDevId:02679: Invalid device header signature : path=/dev/sda4, devVersion=d35a479902627784, sigStart=22eb06be5ea79d49, sigEnd=14963fcafc117152

03/08 13:00:13.039496 0x7f43c9c8eeb0:phyDevMap_ThreadedReadDevId:00736: Failed to read DeviceId of /dev/sda4. rc=351 took 60 milli

03/08 13:00:13.039597 0x7f43c9c6aeb0:mosAsyncIO_CloseFileIntern:00503: Closing device path:/dev/sda6

03/08 13:00:13.039610 0x7f43c9c6aeb0:mosIO_close:00115: Closing FD(24)

03/08 13:00:13.039633 0x7f43c9c6aeb0:phyDev_ReadDevId:02679: Invalid device header signature : path=/dev/sda6, devVersion=178014400000000, sigStart=b18323100000000a, sigEnd=d0970000d09768

03/08 13:00:13.039638 0x7f43c9c6aeb0:phyDevMap_ThreadedReadDevId:00736: Failed to read DeviceId of /dev/sda6. rc=351 took 0 milli

03/08 13:00:17.446133 0x7f43c9c58eb0:mosAsyncIO_DoIOTimedSyncIntern:01336: Time budget exceeded during IO to /dev/sdb (Give-up on inflight IO)

03/08 13:00:17.446152 0x7f43c9c58eb0:mosAsyncIO_WriteZeroTimedSync:00811: mosAsyncIO_DoIOTimedSync returned ERROR, offset = 35651584, writeSize = 1048576

03/08 13:00:17.551561 0x7f43c9c58eb0:mosAsyncIO_DoIOTimedSyncIntern:01336: Time budget exceeded during IO to /dev/sdb (Give-up on inflight IO)

03/08 13:00:17.551570 0x7f43c9c58eb0:mosAsyncIO_WriteZeroTimedSync:00811: mosAsyncIO_DoIOTimedSync returned ERROR, offset = 0, writeSize = 1048576

03/08 13:00:17.551577 0x7f43c9c58eb0:contDev_SendDeviceError:01632: Sending device error to MDM: DevId:ffdff6ca00010000 deviceName: /dev/sdb readError: FALSE WriteError: TRUE

03/08 13:00:17.552437 0x7f43c9c58eb0:phyDev_InitWithBuf:02152: Io error /dev/sdb, size 409600

03/08 13:00:17.552462 0x7f43c9c58eb0:contDev_SendDeviceError:01632: Sending device error to MDM: DevId:ffdff6ca00010000 deviceName: /dev/sdb readError: TRUE WriteError: TRUE

03/08 13:00:17.559601 0x7f43c9c58eb0:contCmd_RemDevIntern:01022: DevId ffdff6ca00010000 - Start rc = SUCCESS

03/08 13:00:17.559623 0x7f43c9c58eb0:contCmd_AddDev:01401: DevId ffdff6ca00010000 - Done rc = TGT_ADD_DEV_IO_ERROR

03/08 13:00:17.559632 0x7f43c9c58eb0:contLia_ChangeLedMode:00424: Light Dev LED rc: TIMEOUT

03/08 13:00:17.559637 0x7f43c8ea2eb0:contCmd_AddDevGroup:03383: Add Dev group done bReconfigInProgress 0, bUpdateDevicePath 1 # of devices 1

03/08 13:00:17.637550 0x7f43c8ea2eb0:contCmd_RemDevIntern:01022: DevId ffdff6ca00010000 - Start rc = SUCCESS

03/08 13:00:17.637559 0x7f43c8ea2eb0:contCmd_RemDevIntern:01044: DevId ffdff6ca00010000 - Done rc = NOT_FOUND

03/08 13:00:17.791942 (nil):mosAsyncIO_CloseFileIntern:00503: Closing device path:/dev/sdb

03/08 13:00:17.791954 (nil):mosIO_close:00115: Closing FD(18)

AGerasimov

8 Posts

August 10th, 2017 16:00

No help with this question?

zkkhan

10 Posts

September 21st, 2017 00:00

Hi RickH,

I am installing the latest release of Scale IO on a Linux Environment(RHEL 7.4) using Virtual Machines. I hit this same issue while installing Scale IO using the Installation Manager. I added the required parameter as suggested in the solution SCI-21795 in the conf file for SDS. After that, I attempted to run the failed command from the installation manger, but it failed again. Following are the logs from the scaleIO.log file for the latest date:

2017-09-21 01:00:42,347 [http-nio-443-exec-9] INFO  c.e.s.s.w.s.ScaleIOSecurityUtils - Authenticating locally

2017-09-21 01:00:47,926 [scheduler-1] INFO  c.e.s.s.i.s.i.CommandQueue - pendingCommands: 1, finishedCommands: 0, archivedCommands: 0, runningCommand: , next command: AddSdsCommand[working on= (MDM) [10.10.51.240]] 40

2017-09-21 01:00:47,928 [scheduler-1] INFO  c.e.s.s.i.s.i.CommandQueue - pendingCommands: 1, finishedCommands: 0, archivedCommands: 0, runningCommand: , next command: AddSdsCommand[working on= (MDM) [10.10.51.240]] 41

2017-09-21 01:00:47,928 [scheduler-1] INFO  c.e.s.s.i.s.i.CommandQueue - pendingCommands: 1, finishedCommands: 0, archivedCommands: 0, runningCommand: , next command: AddSdsCommand[working on= (MDM) [10.10.51.240]] 42

2017-09-21 01:00:52,776 [executor-45] ERROR c.e.e.c.service.CommandService - addSds Sds[node=LinuxSystemNode[nodeName= ,nodeIPs=[10.10.51.242],credentials=NodeCredentials[domain= ,userName=root]],nodeInfo=NodeInfo[diskDevices=[DiskDevice[diskName=/dev/sda2], DiskDevice[diskName=/dev/dm-1], DiskDevice[diskName=/dev/sda], DiskDevice[diskName=/dev/sda1], DiskDevice[diskName=/dev/sr0], DiskDevice[diskName=/dev/dm-0], DiskDevice[diskName=/dev/sdb]],installedComponents=[],timestamp=Tue Sep 19 02:48:27 EDT 2017],sdsName=SDS_10.10.51.242,protectionDomain=default,protectionDomainId=0,faultSet= ,allIPs=[10.10.51.242],sdsOnlyIPs= ,sdcOnlyIPs= ,devices={Device[devicePath=/dev/sdb,storagePool=defaultSP,deviceName= ]},optimized=false,port=7072,useRmCache=false] on 10.10.51.240 failed. ResultCode: TGT_ADD_DEV_IO_ERROR

2017-09-21 01:00:52,800 [executor-44] ERROR c.e.e.c.service.CommandService - addSds Sds[node=LinuxSystemNode[nodeName= ,nodeIPs=[10.10.51.241],credentials=NodeCredentials[domain= ,userName=root]],nodeInfo=NodeInfo[diskDevices=[DiskDevice[diskName=/dev/sda], DiskDevice[diskName=/dev/dm-0], DiskDevice[diskName=/dev/sr0], DiskDevice[diskName=/dev/sdb], DiskDevice[diskName=/dev/sda2], DiskDevice[diskName=/dev/dm-1], DiskDevice[diskName=/dev/sda1]],installedComponents=[],timestamp=Tue Sep 19 02:48:27 EDT 2017],sdsName=SDS_10.10.51.241,protectionDomain=default,protectionDomainId=0,faultSet= ,allIPs=[10.10.51.241],sdsOnlyIPs= ,sdcOnlyIPs= ,devices={Device[devicePath=/dev/sdb,storagePool=defaultSP,deviceName= ]},optimized=false,port=7072,useRmCache=false] on 10.10.51.240 failed. ResultCode: TGT_ADD_DEV_IO_ERROR

2017-09-21 01:00:52,800 [executor-43] ERROR c.e.e.c.service.CommandService - addSds Sds[node=LinuxSystemNode[nodeName= ,nodeIPs=[10.10.51.240],credentials=NodeCredentials[domain= ,userName=root]],nodeInfo=NodeInfo[diskDevices=[DiskDevice[diskName=/dev/dm-1], DiskDevice[diskName=/dev/sda2], DiskDevice[diskName=/dev/sr0], DiskDevice[diskName=/dev/sda], DiskDevice[diskName=/dev/dm-0], DiskDevice[diskName=/dev/sdb], DiskDevice[diskName=/dev/sda1]],installedComponents=[],timestamp=Tue Sep 19 02:48:27 EDT 2017],sdsName=SDS_10.10.51.240,protectionDomain=default,protectionDomainId=0,faultSet= ,allIPs=[10.10.51.240],sdsOnlyIPs= ,sdcOnlyIPs= ,devices={Device[devicePath=/dev/sdb,storagePool=defaultSP,deviceName= ]},optimized=false,port=7072,useRmCache=false] on 10.10.51.240 failed. ResultCode: TGT_ADD_DEV_IO_ERROR

2017-09-21 01:00:52,979 [executor-45] ERROR c.e.s.s.d.i.commands.AddSdsCommand - Error Could not add SDS 10.10.51.242 to Protection Domain default due to: Add SDS device IO error executing command .AddSdsCommand (abort) : com.emc.s3g.scaleio.im.services.installation.configurators.CommandServiceMdmConnection.addSds(CommandServiceMdmConnection.java:1795)

com.emc.s3g.scaleio.domain.installation.commands.AddSdsCommand.executeMdmCommand(AddSdsCommand.java:82)

com.emc.s3g.scaleio.domain.installation.commands.AddSdsCommand.executeMdmCommand(AddSdsCommand.java:40)

com.emc.s3g.scaleio.domain.installation.commands.MdmCommand.executeCommand(MdmCommand.java:107)

com.emc.s3g.scaleio.domain.installation.commands.BaseCommand.call(BaseCommand.java:346)

java.util.concurrent.FutureTask.run(FutureTask.java:266)

java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)

java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)

java.lang.Thread.run(Thread.java:748)

2017-09-21 01:00:52,979 [executor-43] ERROR c.e.s.s.d.i.commands.AddSdsCommand - Error Could not add SDS 10.10.51.240 to Protection Domain default due to: Add SDS device IO error executing command .AddSdsCommand (abort) : com.emc.s3g.scaleio.im.services.installation.configurators.CommandServiceMdmConnection.addSds(CommandServiceMdmConnection.java:1795)

com.emc.s3g.scaleio.domain.installation.commands.AddSdsCommand.executeMdmCommand(AddSdsCommand.java:82)

com.emc.s3g.scaleio.domain.installation.commands.AddSdsCommand.executeMdmCommand(AddSdsCommand.java:40)

com.emc.s3g.scaleio.domain.installation.commands.MdmCommand.executeCommand(MdmCommand.java:107)

com.emc.s3g.scaleio.domain.installation.commands.BaseCommand.call(BaseCommand.java:346)

java.util.concurrent.FutureTask.run(FutureTask.java:266)

java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)

java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)

java.lang.Thread.run(Thread.java:748)

2017-09-21 01:00:52,979 [executor-44] ERROR c.e.s.s.d.i.commands.AddSdsCommand - Error Could not add SDS 10.10.51.241 to Protection Domain default due to: Add SDS device IO error executing command .AddSdsCommand (abort) : com.emc.s3g.scaleio.im.services.installation.configurators.CommandServiceMdmConnection.addSds(CommandServiceMdmConnection.java:1795)

com.emc.s3g.scaleio.domain.installation.commands.AddSdsCommand.executeMdmCommand(AddSdsCommand.java:82)

com.emc.s3g.scaleio.domain.installation.commands.AddSdsCommand.executeMdmCommand(AddSdsCommand.java:40)

com.emc.s3g.scaleio.domain.installation.commands.MdmCommand.executeCommand(MdmCommand.java:107)

com.emc.s3g.scaleio.domain.installation.commands.BaseCommand.call(BaseCommand.java:346)

java.util.concurrent.FutureTask.run(FutureTask.java:266)

java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)

java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)

java.lang.Thread.run(Thread.java:748)

pawelw1

306 Posts

September 21st, 2017 03:00

Hi,

Depending on  your VM configuration and underlying storage it might be hard to get required 50MB/s throughput with vmdks - what value did you set the SCI-21795 parameter to? Can you try to "dd" to /dev/sdb on that VM and see how much are you getting?

You can also try to simply finish the installation and add SDS's manually to the existing cluster.

Cheers,

Pawel

RHasleton1

73 Posts

September 21st, 2017 03:00

A few points here:

1. As Pawel stated, what parameters did you use to add to the SDS?

2. The logs above appear to be from the gateway, during installation. I'm not sure how you would have changed the parameters during the install.

3. The only VMs supported are the SVMs from the VMware template. If this is just for a lab environment, then that is fine.

4. What kind of controller are you using for the virtual disks? paravirtual or LSI? pvscsi would be the recommended option.

5. Because you are using virtual machines, you won't be getting the speed of real hardware. This is most likely where the issue lies.

As Pawel pointed out above, get the SDSes installed and then try adding the disks after the fact and see if that makes a difference.

Good luck,

Rick

zkkhan

10 Posts

September 22nd, 2017 03:00

RickH,

I will be using this setup for demo/lab purpose. I don't think there is any compulsion for SVM template in this case, right?

zkkhan

10 Posts

September 22nd, 2017 03:00

Hi RickH/Pawl

1. I added the following parameter to the SDS config:

                   mdm_to_tgt_net_send_timeout=60

This was what suggest in the solution for SCI-21795.

2. The installation had failed, only then I added the parameter. Following that, I selected the retry option from the Gateway Installer GUI to attempt for adding the SDS devices which failed during the installation.

3. I have deployed the setup on a virtual setup using VMware Esxi 6.5 as the hypervisor. I am not using the SVM template.

4. The controller type for the disk is paravirtual.

Anyhow, I simply removed the parameter and just tried running the failed steps again and it worked. I don't know what was the real issue.

Thanks.

SysEng777

25 Posts

January 12th, 2018 10:00

Since the documentation says mdm_to_tgt_net_send_timeout=60 is in milliseconds, 60 would have made it worse.  If sounds like the default is 4.5 seconds or 4500 ms so set it to something higher than that.

View More

View All

No Events found!

Top