TS-210 crashing on multiple access?

iSCSI related applications
Post Reply
brondi
Starting out
Posts: 16
Joined: Fri Mar 26, 2010 4:10 am

TS-210 crashing on multiple access?

Post by brondi »

Hi All,

I'm trying to use my TS-210 as shared ISCSI disks + quorum in a Sun Cluster configuration (with 2 nodes).The problem I have is that once the Sun machines are configured and try to make reservation on the disks, the whole NAS crashes (red/green status LED + bip). Unfortunately, I can't find any useful log on the NAS so I am not sure of what is happening.

Any Idea? I am happy to provide all the log you need.

Thanks!

UPDATE:
forgot to say, I'm using firmware 3.3.1
User avatar
schumaku
Guru
Posts: 43579
Joined: Mon Jan 21, 2008 4:41 pm
Location: Kloten (Zurich), Switzerland -- Skype: schumaku
Contact:

Re: TS-210 crashing on multiple access?

Post by schumaku »

Since firmware v3.3.0, all QNAP NAS (TS-x10, TS-x19, TS-x39, TS-x59, TS-509, TS-809) share the same iSCSI feature set now., the software specs on the Web are correct, but this change has not been reflected in the comparison table yet.

When I remember right, the Sun Cluster with two nodes and just two connections to the quorum disk used Reserve/Release as per SPC-2, paired with a Sun-invented persistency key on top; with three or more concurrent connections to the quorum makes use of the standards compliant Persistent Group Reservation (PGR) keys as per SPC-3. Somewhat odd, the Sun Cluster docs still speaks of SCSI-2- and SCSI-3, why ever.

Afraid, can't help more for the moment, have no test environment available to set-up and run a Sun Cluster to the TS-x19 here. Let's see if the iSCSI cracks from QNAP can jump in here.

-Kurt.
brondi
Starting out
Posts: 16
Joined: Fri Mar 26, 2010 4:10 am

Re: TS-210 crashing on multiple access?

Post by brondi »

It is exactly as you said. Depending on the number of nodes, Sun Cluster will use the Sun Invented SCSI-2 PGRE (Persistent Group Reservation Emulation), or SCSI-3 PGR.

This is a nice blog article (guess by a Sun Cluster Developper/Engineer?) http://blogs.sun.com/kristien/entry/scs ... un_cluster which explains in detail.

If SCSI3 PGR is supposed to work, I can try (if it is possible) to force to use that..
brondi
Starting out
Posts: 16
Joined: Fri Mar 26, 2010 4:10 am

Re: TS-210 crashing on multiple access?

Post by brondi »

Some more info:

From node 1, I try to add the iscsi quroum:

Code: Select all

bash-3.00# /usr/cluster/bin/clq add d3
Jul 26 22:36:47 netra1 cl_runtime: NOTICE: CMM: Cluster members: netra2 netra1.
Jul 26 22:36:47 netra1 cl_runtime: NOTICE: CMM: node reconfiguration #6 completed.
Jul 26 22:36:50 netra1 cl_runtime: NOTICE: CMM: Quorum device 1 (/dev/did/rdsk/d3s2) added; votecount = 1, bitmask of nodes with configured paths = 0x3.
Jul 26 22:36:50 netra1 cl_runtime: NOTICE: CMM: Registered key on and acquired quorum device 1 (gdevname /dev/did/rdsk/d3s2).
Jul 26 22:36:50 netra1 cl_runtime: NOTICE: CMM: Quorum device /dev/did/rdsk/d3s2: owner set to node 2.

<NAS dies, bip + red/green led>

Jul 26 22:38:09 netra1 iscsi: NOTICE: iscsi connection(7) unable to connect to target iqn.2004-04.com.qnap:ts-210:iscsi.quorum.bef3c1 (errno:145)
Jul 26 22:39:07 netra1 last message repeated 2 times
Jul 26 22:39:13 netra1 iscsi: NOTICE: iscsi connection(7) unable to connect to target iqn.2004-04.com.qnap:ts-210:iscsi.quorum.bef3c1 (errno:146)
bash-3.00# 
On node2:

Code: Select all

Jul 26 22:36:45 netra2 cl_runtime: NOTICE: CMM: Cluster members: netra2 netra1.
Jul 26 22:36:45 netra2 cl_runtime: NOTICE: CMM: node reconfiguration #6 completed.
Jul 26 22:36:49 netra2 cl_runtime: NOTICE: CMM: Quorum device 1 (/dev/did/rdsk/d3s2) added; votecount = 1, bitmask of nodes with configured paths = 0x3.
Jul 26 22:37:43 netra2 scsi: WARNING: /scsi_vhci/ssd@g60014055ecd93fbd4c0cd474cd922dd3 (ssd0):
Jul 26 22:37:43 netra2  SCSI transport failed: reason 'tran_err': retrying command
Jul 26 22:38:07 netra2 iscsi: NOTICE: iscsi connection(7) unable to connect to target iqn.2004-04.com.qnap:ts-210:iscsi.quorum.bef3c1 (errno:145)
Jul 26 22:39:04 netra2 last message repeated 2 times
Jul 26 22:39:11 netra2 iscsi: NOTICE: iscsi connection(7) unable to connect to target iqn.2004-04.com.qnap:ts-210:iscsi.quorum.bef3c1 (errno:146)
Jul 26 22:40:21 netra2 last message repeated 14 times
Jul 26 22:40:26 netra2 cl_runtime: WARNING: CMM: PGRE string missing from quorum device
Jul 26 22:40:26 netra2 cl_runtime: WARNING: QUORUM_GENERIC: quorum_read_keys error: Reading the registration keys failed on quorum device /dev/did/rdsk/d3s2 with error 22.
Jul 26 22:40:26 netra2 cl_runtime: WARNING: CMM: Reading reservation keys from quorum device /dev/did/rdsk/d3s2 failed with error 2.

I've also managed to record the full iSCSI conversation between the 2 nodes and the Qnap (using tshark on the QNAP, so it is recorded up to the crash). I'm happy to send the file (readable with Wireshark) if needed.
QNAPJauss
QNAP Staff
Posts: 499
Joined: Fri Oct 02, 2009 12:18 pm
Location: Taipei, TAIWAN

Re: TS-210 crashing on multiple access?

Post by QNAPJauss »

Dear brondi,

thanks for the feed back.
We will have a look to that problem, and i will keep you inform soon.

Meanwhile, we are interested in the logs you've got.
Check your PM, i will send you the information to send us the logs.

BR,
Jauss
brondi
Starting out
Posts: 16
Joined: Fri Mar 26, 2010 4:10 am

Re: TS-210 crashing on multiple access?

Post by brondi »

I tried to look around for some iscsi log but I can't find any.. Is there a way to start the iscsi target demon to write debug informations?

The system is for personal usage so I don't mind changing/testing/braking things! :-)
brondi
Starting out
Posts: 16
Joined: Fri Mar 26, 2010 4:10 am

Re: TS-210 crashing on multiple access?

Post by brondi »

Got something! I guess it is a kernel bug?

from kmsg:

Code: Select all

<6>TARGET_CORE[iSCSI]: Registered fabric_sess_ptr: cf3f3cc0
<4>iSCSI Login successful on CID: 0 from 192.168.1.220 to 192.168.1.100:3260,1
<4>Incremented iSCSI Connection count to 1 from node: iqn.1986-03.com.sun:01:080020c24764.4c4b4e65
<4>Established iSCSI session from node: iqn.1986-03.com.sun:01:080020c24764.4c4b4e65
<4>Incremented number of active iSCSI sessions to 1 on iSCSI Target Portal Group: 1
<4>Decremented iSCSI connection count to 0 from node: iqn.1986-03.com.sun:01:080020c24764.4c4b4e65
<6>TARGET_CORE[iSCSI]: Deregistered fabric_sess
<4>Released iSCSI session from node: iqn.1986-03.com.sun:01:080020c24764.4c4b4e65
<4>Decremented number of active iSCSI Sessions on iSCSI TPG: 1 to 0
<4>Received iSCSI login request from 192.168.1.220 on TCP Network Portal 192.168.1.100:3260
<4>------------------------------------------------------------------
<4>HeaderDigest:                 None
<4>DataDigest:                   None
<4>MaxRecvDataSegmentLength:     8192
<4>IFMarker:                     No
<4>OFMarker:                     No
<4>------------------------------------------------------------------
<4>------------------------------------------------------------------
<4>InitiatorName:                iqn.1986-03.com.sun:01:080020c24764.4c4b4e65
<4>TargetAlias:                  QNAP Target
<4>InitiatorAlias:               netra2
<4>TargetPortalGroupTag:         1
<4>DefaultTime2Wait:             2
<4>DefaultTime2Retain:           20
<4>ErrorRecoveryLevel:           0
<4>SessionType:                  Discovery
<4>------------------------------------------------------------------
<6>TARGET_CORE[iSCSI]: Registered fabric_sess_ptr: cf3f3180
<4>iSCSI Login successful on CID: 0 from 192.168.1.220 to 192.168.1.100:3260,1
<4>Incremented iSCSI Connection count to 1 from node: iqn.1986-03.com.sun:01:080020c24764.4c4b4e65
<4>Established iSCSI session from node: iqn.1986-03.com.sun:01:080020c24764.4c4b4e65
<4>Incremented number of active iSCSI sessions to 1 on iSCSI Target Portal Group: 1
<4>Decremented iSCSI connection count to 0 from node: iqn.1986-03.com.sun:01:080020c24764.4c4b4e65
<6>TARGET_CORE[iSCSI]: Deregistered fabric_sess
<4>Released iSCSI session from node: iqn.1986-03.com.sun:01:080020c24764.4c4b4e65
<4>Decremented number of active iSCSI Sessions on iSCSI TPG: 1 to 0
<3>vfs_readv() returned -14
<2>kernel BUG at include/linux/scatterlist.h:63!
<1>Unable to handle kernel NULL pointer dereference at virtual address 00000000
<1>pgd = c0004000
<1>[00000000] *pgd=00000000
<4>Internal error: Oops: 817 [#1]
<4>Modules linked in: iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi iscsi_target_mod target_core_mod fbdisk ipv6 nfs udf isofs ufsd(P) ext4 jbd2 sysinfo usblp usbhid usb_storage ahci mv_sata
<4>CPU: 0    Tainted: P            (2.6.30.6 #1)
<4>PC is at __bug+0x20/0x2c
<4>LR is at release_console_sem+0x198/0x1ac
<4>pc : [<c002c98c>]    lr : [<c003f060>]    psr: 20000013
<4>sp : cf0fdde8  ip : cf0fdd18  fp : cf0fddf4
<4>r10: cef88960  r9 : 00000000  r8 : cf0fde94
<4>r7 : 00000000  r6 : 00000000  r5 : 00000001  r4 : cf0fdf0c
<4>r3 : 00000000  r2 : 00000001  r1 : 0000bf87  r0 : 00000031
<4>Flags: nzCv  IRQs on  FIQs on  Mode SVC_32  ISA ARM  Segment kernel
<4>Control: 0005397f  Table: 0f224000  DAC: 00000017
<4>Process QNAP_fileio (pid: 2489, stack limit = 0xcf0fc268)
<4>Stack: (0xcf0fdde8 to 0xcf0fe000)
<4>dde0:                   cf0fde1c cf0fddf8 bf1e7bb0 c002c97c cef88960 cf9e9d14 
<4>de00: cf0fde94 cf0fdf0c cf9e9d20 cf0fde98 cf0fde54 cf0fde20 bf1de4fc bf1e7af4 
<4>de20: cf0fde98 cf0fde94 cf0fdf0c cef889a0 cf05a0c0 bf20e5b4 cf119000 cef88960 
<4>de40: 00000000 00000001 cf0fdecc cf0fde58 bf1e71a8 bf1de458 cf9e9d14 cf0fde98 
<4>de60: cf0fde94 cf0fdf0c 00000001 cf119000 bf20e5b4 cf0fdf38 cf3f5da0 00000001 
<4>de80: cf9e9d14 00000000 00000000 00000000 cf0a0380 00000000 00000000 00000000 
<4>dea0: 00000000 cf3f5da0 bf20e5b4 00000000 000000f1 cf119000 00000000 cf0fc000 
<4>dec0: cf0fdef4 cf0fded0 bf1de360 bf1e6ee0 00000000 00000000 00000001 00000000 
<4>dee0: cf0fdf08 cf0fdf0c cf0fdf2c cf0fdef8 bf1e0ee0 bf1de324 00000001 00000000 
<4>df00: cf0fdf08 cf0fdf0c 00000000 00000000 00000000 cf3f5da0 00000000 cf3f5da0 
<4>df20: cf0fdf94 cf0fdf30 bf1e32bc bf1e0e6c 00000000 cf0a03b0 00000001 00000000 
<4>df40: 00000000 00000000 cf3f5da0 cf119000 cf119000 cf119000 bf20e5b4 bf20e5b4 
<4>df60: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 
<4>df80: cecb39a0 cf3f5da0 cf0fdfdc cf0fdf98 bf1e78bc bf1e3208 cf0fdfc4 cf0fdfa8 
<4>dfa0: 00000000 cecb39a0 c0052534 cf0fdfac cf0fdfac 00000000 cf119000 bf1e735c 
<4>dfc0: 00000000 00000000 00000000 00000000 cf0fdff4 cf0fdfe0 c0052054 bf1e736c 
<4>dfe0: 00000000 00000000 00000000 cf0fdff8 c0041478 c0052008 00000000 00000000 
<4>Backtrace: 
<4>[<c002c96c>] (__bug+0x0/0x2c) from [<bf1e7bb0>] (transport_map_mem_to_sg+0xcc/0x170 [target_core_mod])
<4>[<bf1e7ae4>] (transport_map_mem_to_sg+0x0/0x170 [target_core_mod]) from [<bf1de4fc>] (dev_obj_do_se_mem_map+0xb4/0xbc [target_core_mod])
<4>[<bf1de448>] (dev_obj_do_se_mem_map+0x0/0xbc [target_core_mod]) from [<bf1e71a8>] (transport_generic_get_cdb_count+0x2d8/0x400 [target_core_mod])
<4>[<bf1e6ed0>] (transport_generic_get_cdb_count+0x0/0x400 [target_core_mod]) from [<bf1de360>] (dev_obj_get_cdb_count+0x4c/0x54 [target_core_mod])
<4>[<bf1de314>] (dev_obj_get_cdb_count+0x0/0x54 [target_core_mod]) from [<bf1e0ee0>] (transport_new_cmd_obj+0x84/0x164 [target_core_mod])
<4>[<bf1e0e5c>] (transport_new_cmd_obj+0x0/0x164 [target_core_mod]) from [<bf1e32bc>] (transport_generic_new_cmd+0xc4/0x254 [target_core_mod])
<4> r6:cf3f5da0 r5:00000000 r4:cf3f5da0
<4>[<bf1e31f8>] (transport_generic_new_cmd+0x0/0x254 [target_core_mod]) from [<bf1e78bc>] (transport_processing_thread+0x560/0x788 [target_core_mod])
<4> r6:cf3f5da0 r5:cecb39a0 r4:00000000
<4>[<bf1e735c>] (transport_processing_thread+0x0/0x788 [target_core_mod]) from [<c0052054>] (kthread+0x5c/0x88)
<4>[<c0051ff8>] (kthread+0x0/0x88) from [<c0041478>] (do_exit+0x0/0x5cc)
<4> r5:00000000 r4:00000000
<4>Code: e1a01000 e59f000c eb004b1e e3a03000 (e5833000) 
<4>---[ end trace c6cf7b08edc793d5 ]---
<3>vfs_readv() returned -14
<2>kernel BUG at include/linux/scatterlist.h:63!
<1>Unable to handle kernel NULL pointer dereference at virtual address 00000000
<1>pgd = c0004000
<1>[00000000] *pgd=00000000
<4>Internal error: Oops: 817 [#2]
<4>Modules linked in: iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi iscsi_target_mod target_core_mod fbdisk ipv6 nfs udf isofs ufsd(P) ext4 jbd2 sysinfo usblp usbhid usb_storage ahci mv_sata
<4>CPU: 0    Tainted: P      D     (2.6.30.6 #1)
<4>PC is at __bug+0x20/0x2c
<4>LR is at release_console_sem+0x198/0x1ac
<4>pc : [<c002c98c>]    lr : [<c003f060>]    psr: 20000013
<4>sp : cf1cfde8  ip : cf1cfd18  fp : cf1cfdf4
<4>r10: cef88f00  r9 : 00000000  r8 : cf1cfe94
<4>r7 : 00000000  r6 : 00000000  r5 : 00000002  r4 : cf1cff0c
<4>r3 : 00000000  r2 : 00000001  r1 : 0000ce7b  r0 : 00000031
<4>Flags: nzCv  IRQs on  FIQs on  Mode SVC_32  ISA ARM  Segment kernel
<4>Control: 0005397f  Table: 0f224000  DAC: 00000017
<4>Process QNAP_fileio (pid: 2509, stack limit = 0xcf1ce268)
<4>Stack: (0xcf1cfde8 to 0xcf1d0000)
<4>fde0:                   cf1cfe1c cf1cfdf8 bf1e7bb0 c002c97c cef88f00 cf9e9d74 
<4>fe00: cf1cfe94 cf1cff0c cf9e9d80 cf1cfe98 cf1cfe54 cf1cfe20 bf1de4fc bf1e7af4 
<4>fe20: cf1cfe98 cf1cfe94 cf1cff0c cef88f40 cf05aac0 bf20e5b4 cfac9400 cef88f00 
<4>fe40: 00000000 00000001 cf1cfecc cf1cfe58 bf1e71a8 bf1de458 cf9e9d74 cf1cfe98 
<4>fe60: cf1cfe94 cf1cff0c bf226508 cfac9400 bf20e5b4 cf1cff38 cf3f5be0 00000001 
<4>fe80: cf9e9d74 00000000 00000000 00000000 cf1cfec4 00000000 00000000 00000000 
<4>fea0: cf1cfec4 cf3f5be0 bf20e5b4 00000000 000000f1 cfac9400 00000000 cf1ce000 
<4>fec0: cf1cfef4 cf1cfed0 bf1de360 bf1e6ee0 00000000 00000000 00000001 00000000 
<4>fee0: cf1cff08 cf1cff0c cf1cff2c cf1cfef8 bf1e0ee0 bf1de324 00000001 00000000 
<4>ff00: cf1cff08 cf1cff0c 00000000 00000000 00000000 cf3f5be0 00000000 cf3f5be0 
<4>ff20: cf1cff94 cf1cff30 bf1e32bc bf1e0e6c 00000000 cf0a03b0 00000001 00000000 
<4>ff40: 00000000 00000000 cf3f5be0 cfac9400 cfac9400 cfac9400 bf20e5b4 bf20e5b4 
<4>ff60: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 
<4>ff80: cf06d060 cf3f5be0 cf1cffdc cf1cff98 bf1e78bc bf1e3208 cf1cffc4 cf1cffa8 
<4>ffa0: 00000000 cf06d060 c0052534 cf1cffac cf1cffac 00000000 cfac9400 bf1e735c 
<4>ffc0: 00000000 00000000 00000000 00000000 cf1cfff4 cf1cffe0 c0052054 bf1e736c 
<4>ffe0: 00000000 00000000 00000000 cf1cfff8 c0041478 c0052008 00000000 00000000 
<4>Backtrace: 
<4>[<c002c96c>] (__bug+0x0/0x2c) from [<bf1e7bb0>] (transport_map_mem_to_sg+0xcc/0x170 [target_core_mod])
<4>[<bf1e7ae4>] (transport_map_mem_to_sg+0x0/0x170 [target_core_mod]) from [<bf1de4fc>] (dev_obj_do_se_mem_map+0xb4/0xbc [target_core_mod])
<4>[<bf1de448>] (dev_obj_do_se_mem_map+0x0/0xbc [target_core_mod]) from [<bf1e71a8>] (transport_generic_get_cdb_count+0x2d8/0x400 [target_core_mod])
<4>[<bf1e6ed0>] (transport_generic_get_cdb_count+0x0/0x400 [target_core_mod]) from [<bf1de360>] (dev_obj_get_cdb_count+0x4c/0x54 [target_core_mod])
<4>[<bf1de314>] (dev_obj_get_cdb_count+0x0/0x54 [target_core_mod]) from [<bf1e0ee0>] (transport_new_cmd_obj+0x84/0x164 [target_core_mod])
<4>[<bf1e0e5c>] (transport_new_cmd_obj+0x0/0x164 [target_core_mod]) from [<bf1e32bc>] (transport_generic_new_cmd+0xc4/0x254 [target_core_mod])
<4> r6:cf3f5be0 r5:00000000 r4:cf3f5be0
<4>[<bf1e31f8>] (transport_generic_new_cmd+0x0/0x254 [target_core_mod]) from [<bf1e78bc>] (transport_processing_thread+0x560/0x788 [target_core_mod])
<4> r6:cf3f5be0 r5:cf06d060 r4:00000000
<4>[<bf1e735c>] (transport_processing_thread+0x0/0x788 [target_core_mod]) from [<c0052054>] (kthread+0x5c/0x88)
<4>[<c0051ff8>] (kthread+0x0/0x88) from [<c0041478>] (do_exit+0x0/0x5cc)
<4> r5:00000000 r4:00000000
<4>Code: e1a01000 e59f000c eb004b1e e3a03000 (e5833000) 
<4>---[ end trace c6cf7b08edc793d6 ]---
<3>Got Unknown Mode Page: 0x03
<3>RefCmdSN 0x00000000 does not equal task's CmdSN 0x00000d28. Rejecting ABORT_TASK.
<4>Received iSCSI login request from 192.168.1.210 on TCP Network Portal 192.168.1.100:3260
<4>Located Storage Object: iqn.2004-04.com.qnap:ts-210:iscsi.quorum.bef3c1
<4>Located Portal Group Object: 1
<6>TARGET_CORE[iSCSI]->TPG[1]_LUN[0] - Copying READ-WRITE access for LUN
<6>TARGET_CORE[iSCSI]->TPG[1]_LUN[1] - Copying READ-WRITE access for LUN
<6>TARGET_CORE[iSCSI]->TPG[1]_LUN[2] - Copying READ-WRITE access for LUN
<4>iSCSI_TPG[1] - Added DYNAMIC ACL with TCQ Depth: 16 for iSCSI Initiator Node: iqn.1986-03.com.sun:01:080020da1552.4c4b52aa
<4>Warning: receive 0 bytes in iscsi_target_rx_thread.
<4>Warning: rx_thread(pid 2425 iscsi_trx/1) got signal 6 ignored!!
<3>RefCmdSN 0x00000000 does not equal task's CmdSN 0x000000e5. Rejecting ABORT_TASK.
<4>Warning: receive 0 bytes in iscsi_target_rx_thread.
<4>iscsi_handle_login_thread_timeout:725: ***ERROR*** iSCSI Login timeout on Network Portal 192.168.1.100:3260
<6>device eth0 entered promiscuous mode
<6>device eth0 left promiscuous mode
<6>device eth0 entered promiscuous mode
<6>device eth0 left promiscuous mode
<4>LIO_Target_ConfigFS: DEREGISTER -> iqn.2004-04.com.qnap:ts-210:iscsi.quorum.bef3c1 TPGT: 1 PORTAL: 192.168.1.100:3260
<4>iscsi_log_rcv_msg: get log pid = 6106.
<6>warning: process `pic_raw' used the deprecated sysctl system call with 8.1.2.
<6>warning: process `pic_raw' used the deprecated sysctl system call with 8.1.2.
X read/write support included 
<4>EXT2-fs warning: mounting fs with errors, running e2fsck is recommended
<2>EXT2-fs error (device mtdblock5): ext2_check_page: bad entry in directory #2: : directory entry across blocks - offset=0, inode=4294967295, rec_len=65536, name_len=255
<6>eth0: link up, 1000 Mb/s, full duplex, flow control disabled
<6>kjournald starting.  Commit interval 5 seconds
<6>EXT3 FS on md9, internal journal
<6>EXT3-fs: mounted filesystem with writeback data mode.
<6>md: bind<sda2>             
<3>md: md2: raid array is not clean -- starting background reconstruction
<6>raid1: raid set md2 active with 1 out of 1 mirrors
<6> md2: unknown partition table
<6>Adding 530040k swap on /dev/md2.  Priority:-1 extents:1 across:530040k 
<4>Clear RTC Alarm interrupt. 
<4>Clear RTC Alarm interrupt. 
<4>active port 0 :139         
<4>active port 1 :445         
<4>active port 2 :20          
<4>EXT2-fs warning: mounting fs with errors, running e2fsck is recommended
<2>EXT2-fs error (device mtdblock5): ext2_check_page: bad entry in directory #2: : directory entry across blocks - offset=0, inode=4294967295, rec_len=65536, name_len=255
<6>kjournald starting.  Commit interval 5 seconds
<6>EXT3 FS on sda3, internal journal
<6>EXT3-fs: mounted filesystem with ordered data mode.
<6>eth0: link up, 1000 Mb/s, full duplex, flow control disabled
<6>NET: Registered protocol family 10
<6>lo: Disabled Privacy Extensions
<7>eth0: no IPv6 routers present
<4>active port 0 :139         
<4>active port 1 :445         
<4>active port 2 :20          
<6>warning: `proftpd' uses 32-bit capabilities (legacy support in use)
<4>NFSD: Using /var/lib/nfs/v4recovery as the NFSv4 state recovery directory
<6>NFSD: starting 90-second grace period
<6>TARGET_CORE[0]: Loading Generic Kernel Storage Engine: v3.1.0 on Linux/armv5tel on 2.6.30.6
<6>TARGET_CORE[0]: Initialized ConfigFS Fabric Infrastructure: v2.0.0 on Linux/armv5tel on 2.6.30.6
<6>SE_PC[0] - Registered Plugin Class: TRANSPORT
<6>PLUGIN_TRANSPORT[1] - pscsi registered
<6>PLUGIN_TRANSPORT[2] - stgt registered
<6>CORE_STGT[0]: Bus Initalization complete
<6>PLUGIN_TRANSPORT[4] - iblock registered
<6>PLUGIN_TRANSPORT[5] - rd_dr registered
<6>PLUGIN_TRANSPORT[6] - rd_mcp registered
<6>PLUGIN_TRANSPORT[7] - fileio registered
<6>SE_PC[1] - Registered Plugin Class: OBJ
QNAPJauss
QNAP Staff
Posts: 499
Joined: Fri Oct 02, 2009 12:18 pm
Location: Taipei, TAIWAN

Re: TS-210 crashing on multiple access?

Post by QNAPJauss »

Dear Brondi,
It is exactly as you said. Depending on the number of nodes, Sun Cluster will use the Sun Invented SCSI-2 PGRE (Persistent Group Reservation Emulation), or SCSI-3 PGR.

This is a nice blog article (guess by a Sun Cluster Developper/Engineer?) http://blogs.sun.com/kristien/entry/scs ... un_cluster which explains in detail.

If SCSI3 PGR is supposed to work, I can try (if it is possible) to force to use that..
Thank you for the feedback (again).
What we support is SPC-3 "Persistent Reservation". It's seems that Sun Cluster uses "Persistent Group Reservation", that is not supported.
We will reproduce the environment to study in our lab.

thank you.

BR,
Jauss
brondi
Starting out
Posts: 16
Joined: Fri Mar 26, 2010 4:10 am

Re: TS-210 crashing on multiple access?

Post by brondi »

Hi All,

I managed to find a workaround that allow me to use the QNAP as a quorum device. SunCluster is able to make reservation in another way with this command:

cldevice set -p default_fencing=nofencing d3

After I run this command I can add the ISCSI LUN as a quroum device without any issue.

However, I think it is still important to investigate/resolve the crash: my guess is that the problem is either in the kernel or in the iscsi target stack..
harn61
New here
Posts: 5
Joined: Mon Jun 03, 2019 11:25 pm

Re: TS-210 crashing on multiple access?

Post by harn61 »

Hi,

triggering this rather old email but lets see if i can get my answer here..

From Jauss
"What we support is SPC-3 "Persistent Reservation". It's seems that Sun Cluster uses "Persistent Group Reservation", that is not supported."

is it still the case ie PGR is still not supported?

Cheers
User avatar
dolbyman
Guru
Posts: 35024
Joined: Sat Feb 12, 2011 2:11 am
Location: Vancouver BC , Canada

Re: TS-210 crashing on multiple access?

Post by dolbyman »

no need to dig out a 10 year thread

different firmware and very different hardware used these days

please stick with your thread
viewtopic.php?f=182&t=148809&p=716942#p716942

and open a ticket with qnap
Post Reply

Return to “iSCSI – Target & Virtual Disk”