Re: Infortrend + iSCSI: detected conn error (1011) + TCP Dup ACK
On 11/10/09 11:39 AM, Mike Christie micha...@cs.wisc.edu wrote: What version of open-iscsi were you using and what kernel, and were you using the iscsi kernel modules with open-iscsi.org tarball or from the kernel? iscsi-initiator-utils-6.2.0.871-0.10.el5 kernel-2.6.18-164.2.1.el5 RedHat RPMs It looks like we are sending more IO than the target can handle. In one of those cases it took more than 30 or 60 seconds (depending on your timeout value). What is the value of cat /sys/block/sdXYZ/device/timeout ? If it is 30 or 60 could you increase it to 360? After you login to the target do echo 360 /sys/block/sdXYZ/device/timeout I've tried setting this, but it appears to have no effect - it was 60, and I increased to 360. And what is the value of: iscsiadm -m node -T your_target | grep node.session.cmds_max If that is 128, then could you decrease that to 32 or 16? Run iscsiadm -m node -T your_target -u iscsiadm -m node -T your_target -o update -n node.session.cmds_max -v 32 iscsiad-m node -T your_target -l I've tried setting to both 16 and 32, but it behaves about the same. And if those prevent the io errors then could you do echo noop /sys/block/sdXYZ/queue/scheduler to see if performance increases with a difference scheduler. I really think I'm back to the duplicate ACK problem - see the attached packet dump - at one point there's 30 duplicate ACKs... Interestingly, the storage has worked for the past week - I'm using it as D2D backup. This morning (about 7 days later), it's giving all these duplicate ACKs. I'm currently running into messages such as: Nov 19 09:46:58 backup iscsid: Kernel reported iSCSI connection 2:0 error (1011) state (3) Nov 19 09:47:00 backup kernel: session2: target reset succeeded Nov 19 09:47:01 backup iscsid: connection2:0 is operational after recovery (1 attempts) Nov 19 09:47:10 backup kernel: sd 9:0:0:0: SCSI error: return code = 0x000e Nov 19 09:47:10 backup kernel: end_request: I/O error, dev sdf, sector 8856 Nov 19 09:47:10 backup kernel: device-mapper: multipath: Failing path 8:80. Nov 19 09:47:10 backup kernel: sd 9:0:0:0: SCSI error: return code = 0x000e Nov 19 09:47:10 backup kernel: end_request: I/O error, dev sdf, sector 74424 Nov 19 09:47:10 backup kernel: sd 9:0:0:2: SCSI error: return code = 0x000e Nov 19 09:47:10 backup kernel: end_request: I/O error, dev sdm, sector 8845240 Nov 19 09:47:10 backup kernel: device-mapper: multipath: Failing path 8:192. Nov 19 09:47:10 backup kernel: sd 9:0:0:2: SCSI error: return code = 0x000e Nov 19 09:47:10 backup kernel: end_request: I/O error, dev sdm, sector 62915456 Nov 19 09:47:10 backup kernel: sd 9:0:0:2: timing out command, waited 300s Nov 19 09:47:10 backup multipathd: /sbin/mpath_prio_alua exitted with 1 Nov 19 09:47:10 backup multipathd: error calling out /sbin/mpath_prio_alua /dev/sdm Nov 19 09:47:10 backup multipathd: 3600d0230061d4479bfb83902: switch to path group #2 This is also interesting: Nov 18 01:48:30 backup multipathd: 3600d0230061d4479bfb83902: remaining active paths: 8 Nov 18 20:16:29 backup multipathd: 3600d0230061d4479bfb83902: remaining active paths: 7 Nov 18 20:16:29 backup multipathd: 3600d0230061d4479bfb83902: remaining active paths: 6 Nov 18 20:16:29 backup multipathd: 3600d0230061d4479bfb83902: remaining active paths: 5 Nov 18 20:16:29 backup multipathd: 3600d0230061d4479bfb83902: remaining active paths: 4 Nov 18 20:16:34 backup multipathd: 3600d0230061d4479bfb83902: remaining active paths: 5 Nov 18 20:32:09 backup multipathd: 3600d0230061d4479bfb83902: remaining active paths: 6 Nov 18 20:43:05 backup multipathd: 3600d0230061d4479bfb83902: remaining active paths: 7 Nov 18 20:48:08 backup multipathd: 3600d0230061d4479bfb83902: remaining active paths: 8 Nov 18 20:53:36 backup multipathd: 3600d0230061d4479bfb83902: remaining active paths: 7 Nov 18 20:53:36 backup multipathd: 3600d0230061d4479bfb83902: remaining active paths: 6 Nov 18 20:53:36 backup multipathd: 3600d0230061d4479bfb83902: remaining active paths: 5 Nov 18 20:53:36 backup multipathd: 3600d0230061d4479bfb83902: remaining active paths: 4 Nov 18 20:53:36 backup multipathd: 3600d0230061d4479bfb83902: remaining active paths: 3 Nov 18 20:53:36 backup multipathd: 3600d0230061d4479bfb83902: remaining active paths: 2 Nov 18 20:53:36 backup multipathd: 3600d0230061d4479bfb83902: remaining active paths: 1 Nov 18 20:53:36 backup multipathd: 3600d0230061d4479bfb83902: remaining active paths: 0 Nov 18 20:53:41 backup multipathd: 3600d0230061d4479bfb83902: remaining active paths: 1 Nov 18 20:59:09 backup multipathd: 3600d0230061d4479bfb83902: remaining active paths: 2 Nov 18 21:04:37 backup multipathd: 3600d0230061d4479bfb83902: remaining active paths: 3 Nov 18 21:10:05 backup multipathd:
Re: Infortrend + iSCSI: detected conn error (1011) + TCP Dup ACK
Matthew Dickinson wrote: On 11/10/09 11:39 AM, Mike Christie micha...@cs.wisc.edu wrote: What version of open-iscsi were you using and what kernel, and were you using the iscsi kernel modules with open-iscsi.org tarball or from the kernel? iscsi-initiator-utils-6.2.0.871-0.10.el5 kernel-2.6.18-164.2.1.el5 RedHat RPMs It looks like we are sending more IO than the target can handle. In one of those cases it took more than 30 or 60 seconds (depending on your timeout value). What is the value of cat /sys/block/sdXYZ/device/timeout ? If it is 30 or 60 could you increase it to 360? After you login to the target do echo 360 /sys/block/sdXYZ/device/timeout I've tried setting this, but it appears to have no effect - it was 60, and I increased to 360. And what is the value of: iscsiadm -m node -T your_target | grep node.session.cmds_max If that is 128, then could you decrease that to 32 or 16? Run iscsiadm -m node -T your_target -u iscsiadm -m node -T your_target -o update -n node.session.cmds_max -v 32 iscsiad-m node -T your_target -l I've tried setting to both 16 and 32, but it behaves about the same. And if those prevent the io errors then could you do echo noop /sys/block/sdXYZ/queue/scheduler to see if performance increases with a difference scheduler. I really think I'm back to the duplicate ACK problem - see the attached packet dump - at one point there's 30 duplicate ACKs... Interestingly, the I did not get the attachement. storage has worked for the past week - I'm using it as D2D backup. This morning (about 7 days later), it's giving all these duplicate ACKs. I'm currently running into messages such as: Nov 19 09:46:58 backup iscsid: Kernel reported iSCSI connection 2:0 error (1011) state (3) Nov 19 09:47:00 backup kernel: session2: target reset succeeded If you are using Red Hat RPMs, make a red hat bugzilla https://bugzilla.redhat.com/. CC mchri...@redhat.com on the bugzilla or email me at that address when you have made the bugzilla. I will then add some network people to it. Attach your trace to the bugzilla. -- You received this message because you are subscribed to the Google Groups open-iscsi group. To post to this group, send email to open-is...@googlegroups.com. To unsubscribe from this group, send email to open-iscsi+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/open-iscsi?hl=.
Re: Infortrend + iSCSI: detected conn error (1011) + TCP Dup ACK
sorry... wrong information. Here is the correct information. I was doing some testing in VMWare Fusion VM's for a presentation that I'm giving. The storage server is CentOS 5.3, which dishes out IETD targets for my OVM servers. The OVM 2.2 environment is as follows: [r...@ovm1 ~]# uname -r 2.6.18-128.2.1.4.9.el5xen [r...@ovm1 ~]# rpm -qa | grep iscsi iscsi-initiator-utils-6.2.0.871-0.7.el5 [r...@ovm1 ~]# On Nov 10, 2009, at 2:30 PM, Mike Christie wrote: Hoot, Joseph wrote: [r...@storage ~]# uname -r 2.6.18-164.el5 [r...@storage ~]# rpm -qa | grep iscsi iscsi-initiator-utils-6.2.0.868-0.18.el5_3.1 [r...@storage ~]# Weird. Is 2.6.18-164.el5 the kernel being used in the virtual machine/DonU? Is that where you are using iscsi? It looks like the Oracle enterprise linux kernel is 2.6.18-164.el5, which looks like it is based on RHEL 5.4. The iscsi code in there is the same as RHEL/upstream. No sendwait patch. However, it looks like there is a 2.6.18-128.2.1.4.9 kernel (comes with the Oracle VM rpms). In here we have a different iscsi version. It looks a little older than what is in 2.6.18-164.el5, but it has the sendwait patch I send to dell. Do you use this kernel in the Dom0? Are you using this kernel with iscsi? On Nov 10, 2009, at 12:17 PM, Mike Christie wrote: Hoot, Joseph wrote: I've had about 3 threads of dt (kicking off a bit randomly) on (3) separate volumes for over a week and haven't had a single disconnect yet. I am currently using whatever rpm is distributed with Oracle VM v2.2. I know for sure that they have included the 871 base, plus I believe at least a one off patch. I can get more details if you'd like. But so far so good for now I think I have the source they are using. Could you do a uname -r, so I can see what kernel they are using. === Joseph R. Hoot Lead System Programmer/Analyst (w) 716-878-4832 (c) 716-759-HOOT joe.h...@itec.suny.edu GPG KEY: 7145F633 === === Joseph R. Hoot Lead System Programmer/Analyst (w) 716-878-4832 (c) 716-759-HOOT joe.h...@itec.suny.edu GPG KEY: 7145F633 === --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups open-iscsi group. To post to this group, send email to open-iscsi@googlegroups.com To unsubscribe from this group, send email to open-iscsi+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/open-iscsi -~--~~~~--~~--~--~---
Re: Infortrend + iSCSI: detected conn error (1011) + TCP Dup ACK
Hoot, Joseph wrote: I've had about 3 threads of dt (kicking off a bit randomly) on (3) separate volumes for over a week and haven't had a single disconnect yet. I am currently using whatever rpm is distributed with Oracle VM v2.2. I know for sure that they have included the 871 base, plus I believe at least a one off patch. I can get more details if you'd like. But so far so good for now I think I have the source they are using. Could you do a uname -r, so I can see what kernel they are using. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups open-iscsi group. To post to this group, send email to open-iscsi@googlegroups.com To unsubscribe from this group, send email to open-iscsi+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/open-iscsi -~--~~~~--~~--~--~---
Re: Infortrend + iSCSI: detected conn error (1011) + TCP Dup ACK
[r...@storage ~]# uname -r 2.6.18-164.el5 [r...@storage ~]# rpm -qa | grep iscsi iscsi-initiator-utils-6.2.0.868-0.18.el5_3.1 [r...@storage ~]# On Nov 10, 2009, at 12:17 PM, Mike Christie wrote: Hoot, Joseph wrote: I've had about 3 threads of dt (kicking off a bit randomly) on (3) separate volumes for over a week and haven't had a single disconnect yet. I am currently using whatever rpm is distributed with Oracle VM v2.2. I know for sure that they have included the 871 base, plus I believe at least a one off patch. I can get more details if you'd like. But so far so good for now I think I have the source they are using. Could you do a uname -r, so I can see what kernel they are using. === Joseph R. Hoot Lead System Programmer/Analyst (w) 716-878-4832 (c) 716-759-HOOT joe.h...@itec.suny.edu GPG KEY: 7145F633 === --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups open-iscsi group. To post to this group, send email to open-iscsi@googlegroups.com To unsubscribe from this group, send email to open-iscsi+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/open-iscsi -~--~~~~--~~--~--~---
Re: Infortrend + iSCSI: detected conn error (1011) + TCP Dup ACK
Matthew Dickinson wrote: On 11/6/09 3:39 PM, Matthew Dickinson matt-openis...@alpha345.com wrote: Try disabling nops by setting node.conn[0].timeo.noop_out_interval = 0 node.conn[0].timeo.noop_out_timeout = 0 I'm still getting errors: Nov 10 09:08:04 backup kernel: connection12:0: detected conn error (1011) Nov 10 09:08:05 backup iscsid: Kernel reported iSCSI connection 12:0 error (1011) state (3) Nov 10 09:08:08 backup iscsid: connection12:0 is operational after recovery (1 attempts) Nov 10 09:09:43 backup kernel: connection11:0: detected conn error (1011) Nov 10 09:09:43 backup kernel: connection12:0: detected conn error (1011) Nov 10 09:09:44 backup kernel: connection11:0: detected conn error (1011) Nov 10 09:09:44 backup iscsid: Kernel reported iSCSI connection 11:0 error (1011) state (3) Nov 10 09:09:44 backup iscsid: Kernel reported iSCSI connection 12:0 error (1011) state (3) Nov 10 09:09:44 backup iscsid: Kernel reported iSCSI connection 11:0 error (1011) state (1) Nov 10 09:09:46 backup kernel: session11: target reset succeeded\ Nov 10 09:09:47 backup iscsid: connection11:0 is operational after recovery (1 attempts) Nov 10 09:09:47 backup iscsid: connection12:0 is operational after recovery (1 attempts) Nov 10 09:09:56 backup kernel: sd 18:0:0:2: SCSI error: return code = 0x000e Nov 10 09:09:56 backup kernel: end_request: I/O error, dev sdv, sector 60721248 Nov 10 09:09:56 backup kernel: device-mapper: multipath: Failing path 65:80. Nov 10 09:09:56 backup kernel: sd 18:0:0:2: SCSI error: return code = 0x000e Nov 10 09:09:56 backup kernel: end_request: I/O error, dev sdv, sector 60727648 Nov 10 09:09:56 backup kernel: sd 18:0:0:2: SCSI error: return code = 0x000e Nov 10 09:10:31 backup kernel: device-mapper: multipath: Failing path 65:112. Interestingly, I tried a Windows 2008 server R2 talking over a single connection to the storage unit, configured to access just via one interface, I was able to sustain 20MB/s so it would ³appear² to be a Linux-related issue - I'm only able to get 9MB/s out of Linux even when using 8 interfaces on both controllers. What version of open-iscsi were you using and what kernel, and were you using the iscsi kernel modules with open-iscsi.org tarball or from the kernel? It looks like we are sending more IO than the target can handle. In one of those cases it took more than 30 or 60 seconds (depending on your timeout value). What is the value of cat /sys/block/sdXYZ/device/timeout ? If it is 30 or 60 could you increase it to 360? After you login to the target do echo 360 /sys/block/sdXYZ/device/timeout And what is the value of: iscsiadm -m node -T your_target | grep node.session.cmds_max If that is 128, then could you decrease that to 32 or 16? Run iscsiadm -m node -T your_target -u iscsiadm -m node -T your_target -o update -n node.session.cmds_max -v 32 iscsiad-m node -T your_target -l And if those prevent the io errors then could you do echo noop /sys/block/sdXYZ/queue/scheduler to see if performance increases with a difference scheduler. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups open-iscsi group. To post to this group, send email to open-iscsi@googlegroups.com To unsubscribe from this group, send email to open-iscsi+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/open-iscsi -~--~~~~--~~--~--~---
Re: Infortrend + iSCSI: detected conn error (1011) + TCP Dup ACK
Hoot, Joseph wrote: [r...@storage ~]# uname -r 2.6.18-164.el5 [r...@storage ~]# rpm -qa | grep iscsi iscsi-initiator-utils-6.2.0.868-0.18.el5_3.1 [r...@storage ~]# Weird. Is 2.6.18-164.el5 the kernel being used in the virtual machine/DonU? Is that where you are using iscsi? It looks like the Oracle enterprise linux kernel is 2.6.18-164.el5, which looks like it is based on RHEL 5.4. The iscsi code in there is the same as RHEL/upstream. No sendwait patch. However, it looks like there is a 2.6.18-128.2.1.4.9 kernel (comes with the Oracle VM rpms). In here we have a different iscsi version. It looks a little older than what is in 2.6.18-164.el5, but it has the sendwait patch I send to dell. Do you use this kernel in the Dom0? Are you using this kernel with iscsi? On Nov 10, 2009, at 12:17 PM, Mike Christie wrote: Hoot, Joseph wrote: I've had about 3 threads of dt (kicking off a bit randomly) on (3) separate volumes for over a week and haven't had a single disconnect yet. I am currently using whatever rpm is distributed with Oracle VM v2.2. I know for sure that they have included the 871 base, plus I believe at least a one off patch. I can get more details if you'd like. But so far so good for now I think I have the source they are using. Could you do a uname -r, so I can see what kernel they are using. === Joseph R. Hoot Lead System Programmer/Analyst (w) 716-878-4832 (c) 716-759-HOOT joe.h...@itec.suny.edu GPG KEY: 7145F633 === --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups open-iscsi group. To post to this group, send email to open-iscsi@googlegroups.com To unsubscribe from this group, send email to open-iscsi+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/open-iscsi -~--~~~~--~~--~--~---
Re: Infortrend + iSCSI: detected conn error (1011) + TCP Dup ACK
El 06/11/09 14:10, mdaitc escribió: Hi mdaitc, I’m seeing similar TCP “weirdness” as the other posts mention as well as the below errors. (..) Nov 2 08:15:14 backup kernel: connection33:0: detected conn error The performance isn’t what I’d expect: (..) What happens if you disable TCP window scaling option in RHEL servers? # echo 0 /proc/sys/net/ipv4/tcp_window_scaling In our case, iSCSI conn errors stopped after disabling, but still have a lot of TCP “weirdness” in the network, mainly dup ACKs packages. Regards, -- Santi Saez http://woop.es --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups open-iscsi group. To post to this group, send email to open-iscsi@googlegroups.com To unsubscribe from this group, send email to open-iscsi+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/open-iscsi -~--~~~~--~~--~--~---
Re: Infortrend + iSCSI: detected conn error (1011) + TCP Dup ACK
What version of OiS are you using? I had lots of weirdness and the same types of disconnects to our Dell EqualLogic when we were (actually still are in production) using 868 code. I'm now using open- iscsi-871 code plus a sendwait patch and haven' had the issue. I've now been slamming my storage for a week and a half with multiple threads of dt. On Nov 9, 2009, at 4:33 AM, Santi Saez wrote: El 06/11/09 14:10, mdaitc escribió: Hi mdaitc, I’m seeing similar TCP “weirdness” as the other posts mention as well as the below errors. (..) Nov 2 08:15:14 backup kernel: connection33:0: detected conn error The performance isn’t what I’d expect: (..) What happens if you disable TCP window scaling option in RHEL servers? # echo 0 /proc/sys/net/ipv4/tcp_window_scaling In our case, iSCSI conn errors stopped after disabling, but still have a lot of TCP “weirdness” in the network, mainly dup ACKs packages. Regards, -- Santi Saez http://woop.es === Joseph R. Hoot Lead System Programmer/Analyst (w) 716-878-4832 (c) 716-759-HOOT joe.h...@itec.suny.edu GPG KEY: 7145F633 === --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups open-iscsi group. To post to this group, send email to open-iscsi@googlegroups.com To unsubscribe from this group, send email to open-iscsi+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/open-iscsi -~--~~~~--~~--~--~---
Re: Infortrend + iSCSI: detected conn error (1011) + TCP Dup ACK
Hi all, I am working on iSCSI En. Tar. Could you please someone explain about the performance of the IET. If so how the performance was calculated and what was the througput for the same. Thanks Gopala krishnan Varatharajan On Sat, Nov 7, 2009 at 3:09 AM, Matthew Dickinson matt-openis...@alpha345.com wrote: On 11/6/09 3:08 PM, Mike Christie micha...@cs.wisc.edu wrote: Could you send more of the log? Do you see a message like connection1:0 is operational after recovery (1 attempts) after you see the conn errors (how many attempts)? Here's one particular connection: Nov 4 05:12:14 backup kernel: connection22:0: ping timeout of 5 secs expired, recv timeout 5, last rx 4321648393, last ping 4321653393, now 4321658393 Nov 4 05:12:14 backup kernel: connection22:0: detected conn error (1011) Nov 4 05:12:21 backup iscsid: connection22:0 is operational after recovery (1 attempts) Nov 4 05:12:46 backup kernel: connection22:0: ping timeout of 5 secs expired, recv timeout 5, last rx 4321680691, last ping 4321685691, now 4321690691 Nov 4 05:12:46 backup kernel: connection22:0: detected conn error (1011) Nov 4 05:12:58 backup iscsid: connection22:0 is operational after recovery (1 attempts) Nov 4 07:46:03 backup kernel: connection22:0: ping timeout of 5 secs expired, recv timeout 5, last rx 4330877890, last ping 4330882890, now 4330887890 Nov 4 07:46:03 backup kernel: connection22:0: detected conn error (1011) Nov 4 07:46:10 backup iscsid: connection22:0 is operational after recovery (1 attempts) Nov 4 07:46:27 backup kernel: connection22:0: ping timeout of 5 secs expired, recv timeout 5, last rx 4330901733, last ping 4330906733, now 4330911733 Nov 4 07:46:27 backup kernel: connection22:0: detected conn error (1011) Nov 4 07:46:32 backup iscsid: connection22:0 is operational after recovery (1 attempts) Nov 4 07:47:21 backup kernel: connection22:0: ping timeout of 5 secs expired, recv timeout 5, last rx 4330955414, last ping 4330960414, now 4330965414 Nov 4 07:47:21 backup kernel: connection22:0: detected conn error (1011) Nov 4 07:47:28 backup iscsid: connection22:0 is operational after recovery (1 attempts) Nov 4 07:48:28 backup kernel: connection22:0: ping timeout of 5 secs expired, recv timeout 5, last rx 4331023213, last ping 4331028213, now 4331033213 Nov 4 07:48:28 backup kernel: connection22:0: detected conn error (1011) Nov 4 07:48:35 backup iscsid: connection22:0 is operational after recovery (1 attempts) FWIW: [r...@backup ~]# cat /var/log/messages | grep after recovery | awk '{print $11 $12}' | sort | uniq (113 attempts) (1 attempts) (24 attempts) (2 attempts) (3 attempts) (4 attempts) (5 attempts) (66 attempts) (68 attempts) (6 attempts) (7 attempts) (8 attempts) (9 attempts) Try disabling nops by setting node.conn[0].timeo.noop_out_interval = 0 node.conn[0].timeo.noop_out_timeout = 0 Ok, I'll let you know how it pans out. Thanks, Matthew -- --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups open-iscsi group. To post to this group, send email to open-iscsi@googlegroups.com To unsubscribe from this group, send email to open-iscsi+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/open-iscsi -~--~~~~--~~--~--~---
Re: Infortrend + iSCSI: detected conn error (1011) + TCP Dup ACK
Hoot, Joseph wrote: What version of OiS are you using? I had lots of weirdness and the same types of disconnects to our Dell EqualLogic when we were (actually still are in production) using 868 code. I'm now using open- iscsi-871 code plus a sendwait patch and haven' had the issue. I've What is the sendwait patch? Is it a patch for open-iscsi or to the kernel network code? --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups open-iscsi group. To post to this group, send email to open-iscsi@googlegroups.com To unsubscribe from this group, send email to open-iscsi+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/open-iscsi -~--~~~~--~~--~--~---
Re: Infortrend + iSCSI: detected conn error (1011) + TCP Dup ACK
it was for OiS 871 code prior to RHEL 5.4 release (not sure if the release include it or not). I'm not sure who came up with it. I was working with Don Williams from Dell EqualLogic. He got ahold of it somehow. I applied it and it seemed to improve things. On Nov 9, 2009, at 2:31 PM, Mike Christie wrote: Hoot, Joseph wrote: What version of OiS are you using? I had lots of weirdness and the same types of disconnects to our Dell EqualLogic when we were (actually still are in production) using 868 code. I'm now using open- iscsi-871 code plus a sendwait patch and haven' had the issue. I've What is the sendwait patch? Is it a patch for open-iscsi or to the kernel network code? === Joseph R. Hoot Lead System Programmer/Analyst (w) 716-878-4832 (c) 716-759-HOOT joe.h...@itec.suny.edu GPG KEY: 7145F633 === --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups open-iscsi group. To post to this group, send email to open-iscsi@googlegroups.com To unsubscribe from this group, send email to open-iscsi+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/open-iscsi -~--~~~~--~~--~--~---
Re: Infortrend + iSCSI: detected conn error (1011) + TCP Dup ACK
Hoot, Joseph wrote: it was for OiS 871 code prior to RHEL 5.4 release (not sure if the release include it or not). I'm not sure who came up with it. I was working with Don Williams from Dell EqualLogic. He got ahold of it somehow. I applied it and it seemed to improve things. Ah ok. I think it was the patch I sent to Don. If you just used 871 without the patch (or what is in the stock RHEL 5.4 kernel) does it work ok? There were a couple changes from 868 to 871 that I thought would also fix the problem, so I was waiting for Don and them to retest just 871 and get back to me. On Nov 9, 2009, at 2:31 PM, Mike Christie wrote: Hoot, Joseph wrote: What version of OiS are you using? I had lots of weirdness and the same types of disconnects to our Dell EqualLogic when we were (actually still are in production) using 868 code. I'm now using open- iscsi-871 code plus a sendwait patch and haven' had the issue. I've What is the sendwait patch? Is it a patch for open-iscsi or to the kernel network code? === Joseph R. Hoot Lead System Programmer/Analyst (w) 716-878-4832 (c) 716-759-HOOT joe.h...@itec.suny.edu GPG KEY: 7145F633 === --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups open-iscsi group. To post to this group, send email to open-iscsi@googlegroups.com To unsubscribe from this group, send email to open-iscsi+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/open-iscsi -~--~~~~--~~--~--~---
Re: Infortrend + iSCSI: detected conn error (1011) + TCP Dup ACK
I've had about 3 threads of dt (kicking off a bit randomly) on (3) separate volumes for over a week and haven't had a single disconnect yet. I am currently using whatever rpm is distributed with Oracle VM v2.2. I know for sure that they have included the 871 base, plus I believe at least a one off patch. I can get more details if you'd like. But so far so good for now On Nov 9, 2009, at 6:18 PM, Mike Christie wrote: Hoot, Joseph wrote: it was for OiS 871 code prior to RHEL 5.4 release (not sure if the release include it or not). I'm not sure who came up with it. I was working with Don Williams from Dell EqualLogic. He got ahold of it somehow. I applied it and it seemed to improve things. Ah ok. I think it was the patch I sent to Don. If you just used 871 without the patch (or what is in the stock RHEL 5.4 kernel) does it work ok? There were a couple changes from 868 to 871 that I thought would also fix the problem, so I was waiting for Don and them to retest just 871 and get back to me. On Nov 9, 2009, at 2:31 PM, Mike Christie wrote: Hoot, Joseph wrote: What version of OiS are you using? I had lots of weirdness and the same types of disconnects to our Dell EqualLogic when we were (actually still are in production) using 868 code. I'm now using open- iscsi-871 code plus a sendwait patch and haven' had the issue. I've What is the sendwait patch? Is it a patch for open-iscsi or to the kernel network code? === Joseph R. Hoot Lead System Programmer/Analyst (w) 716-878-4832 (c) 716-759-HOOT joe.h...@itec.suny.edu GPG KEY: 7145F633 === === Joseph R. Hoot Lead System Programmer/Analyst (w) 716-878-4832 (c) 716-759-HOOT joe.h...@itec.suny.edu GPG KEY: 7145F633 === --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups open-iscsi group. To post to this group, send email to open-iscsi@googlegroups.com To unsubscribe from this group, send email to open-iscsi+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/open-iscsi -~--~~~~--~~--~--~---
Re: Infortrend + iSCSI: detected conn error (1011) + TCP Dup ACK
El 03/11/09 0:52, Mike Christie escribió: Dear Mike, You can turn off ping/nops by setting node.conn[0].timeo.noop_out_interval = 0 node.conn[0].timeo.noop_out_timeout = 0 (set that in iscsid.conf then rediscovery the target or run iscsiadm -m node -T your_target -o update -n name_of_param_above -v 0 Thanks!! As I said to James in the previous email, disabling TCP window scaling *solves partially* this problem, we still hold nop pings in the configuration. But still have too many TCP Dup ACKs in the network :-S This might just work around. What might happen is that you will not see the nop/ping and conn errors and instead would just see a slow down in the workloads being run. I have sent your contact to Infortrend developers, a engineer will contact you, thanks! Regards, -- Santi Saez http://woop.es --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups open-iscsi group. To post to this group, send email to open-iscsi@googlegroups.com To unsubscribe from this group, send email to open-iscsi+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/open-iscsi -~--~~~~--~~--~--~---
Re: Infortrend + iSCSI: detected conn error (1011) + TCP Dup ACK
Santi Saez wrote: Hi, Randomly we get Open-iSCSI conn errors when connecting to an Infortrend A16E-G2130-4 storage array. We had discussed about this earlier in the list, see: http://tr.im/DVQm http://tr.im/DVQp Open-iSCSI logs this: === Nov 2 18:34:02 vz-17 kernel: ping timeout of 5 secs expired, last rx 408250499, last ping 408249467, now 408254467 Nov 2 18:34:02 vz-17 kernel: connection1:0: iscsi: detected conn error (1011) Nov 2 18:34:03 vz-17 iscsid: Kernel reported iSCSI connection 1:0 error (1011) state (3) Nov 2 18:34:07 vz-17 iscsid: connection1:0 is operational after recovery (1 attempts) Nov 2 18:34:52 vz-17 kernel: ping timeout of 5 secs expired, last rx 408294833, last ping 408299833, now 408304833 Nov 2 18:34:52 vz-17 kernel: connection1:0: iscsi: detected conn error (1011) Nov 2 18:34:53 vz-17 iscsid: Kernel reported iSCSI connection 1:0 error (1011) state (3) Nov 2 18:34:57 vz-17 iscsid: connection1:0 is operational after recovery (1 attempts) === Running on CentOS 5.4 with iscsi-initiator-utils-6.2.0.871-0.10.el5; I think it's not a Open-iSCSI bug as Mike suggested at: http://groups.google.com/group/open-iscsi/msg/fe37156096b2955f I have only this error when connecting to Infortrend storage, and not with NetApp, Nexsan, etc. *connected in the same SAN*. Using Wireshark I see a lot of TCP Dup ACK, TCP ACKed lost segment, etc. and iSCSI session finally ends in timeout, see a screenshot here: http://tinyurl.com/ykpvckn Using Wireshark IO graphs I get this strange report about TCP/IP errors: http://tinyurl.com/ybm4m8x And this is another report in the same SAN connecting to a NetApp: http://tinyurl.com/ycgc8ul Those TCP/IP errors only occurs when connecting to Infortrend storage.. and no with other targets in the same SAN (using same switch infrastructure); is there anyway to deal with this using Open-iSCSI? As I see in Internet, there're a lot of Infortrend's users suffering this behavior. You can turn off ping/nops by setting node.conn[0].timeo.noop_out_interval = 0 node.conn[0].timeo.noop_out_timeout = 0 (set that in iscsid.conf then rediscovery the target or run iscsiadm -m node -T your_target -o update -n name_of_param_above -v 0 Or you might want to set them higher. This might just work around. What might happen is that you will not see the nop/ping and conn errors and instead would just see a slow down in the workloads being run. If you guys can get a hold of any infrotrend people let me know, because I would be happy to work with them on this. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups open-iscsi group. To post to this group, send email to open-iscsi@googlegroups.com To unsubscribe from this group, send email to open-iscsi+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/open-iscsi -~--~~~~--~~--~--~---
Re: Infortrend + iSCSI: detected conn error (1011) + TCP Dup ACK
Hi! I would guess that your network/storage system is overloaded occasionally, and then network packets are significantly (5s) delayed. It sounds unlikely, but that would explain a duplicate ACK IMHO. Here with SLES10 SP2 on x86_64 with a HP EVA 6000 + iSCSI connectivity option (MPX100) there is no such problem, even though there are about 20 hosts connected to the storage system, most of them directly through a fabric switch. However the iSCSI load is light. Here is what one of the two MPX boxes reports (the load will rise; it's still early morning): # show perf byte WARNING: Valid data is only displayed for port(s) that are not associated with any configured FCIP routes. Displaying bytes/sec (total)... (Press any key to stop display) GE1 GE2 FC1 FC2 53K 606K43K 620K 774 774 1K 1K 651K868 17K 636K 2K 26K 18K 10K 49K 0 22K 26K 55K 805K799K66K 29K 17K 40K 7K 597K7K 534K72K 172 196K150K47K 49K 516 25K 28K 294K43K 29K 313K 774 774 1K 1K 228K868 65K 171K 18K 20K 528 43K 6K 6K 0 12K 145K516 101K45K 15K 29K 15K 30K (GEx are the gigabit Ethernet ports, FCx are the FibreChannel ports that are connected with the storage system via a fabric switch) Some more statistics (for those who care or like inspiration): # show stats FC Port Statistics FC Port 1 Interrupt Count 82812963 Target Command Count 0 Initiator Command Count 83526964 Link Failure Count 0 Loss of Sync Count 0 Loss of Signal Count 0 Primitive Sequence Error Count 0 Invalid Transmission Word Count 0 Invalid CRC Error Count 0 FC Port 2 Interrupt Count 81551444 Target Command Count 0 Initiator Command Count 82221649 Link Failure Count 0 Loss of Sync Count 0 Loss of Signal Count 0 Primitive Sequence Error Count 0 Invalid Transmission Word Count 0 Invalid CRC Error Count 0 iSCSI Port Statistics --- iSCSI Port 1 Interrupt Count 190227478 Target Command Count 100974368 Initiator Command Count 0 MAC Xmit Frames 122882662 MAC Xmit Byte Count 28445475500 MAC Xmit Multicast Frames0 MAC Xmit Broadcast Frames0 MAC Xmit Pause Frames0 MAC Xmit Control Frames 0 MAC Xmit Deferrals 0 MAC Xmit Late Collisions 0 MAC Xmit Aborted 0 MAC Xmit Single Collisions 0 MAC Xmit Multiple Collisions 0 MAC Xmit Collisions 0 MAC Xmit Dropped Frames 0 MAC Xmit Jumbo Frames1702770 MAC Rcvd Frames 156585933 MAC Rcvd Byte Count 187107689124 MAC Rcvd Unknown Control Frames 0 MAC Rcvd Pause Frames0 MAC Rcvd Control Frames 0 MAC Rcvd Dribbles0 MAC Rcvd Frame Length Errors 0 MAC Rcvd Jabbers 0 MAC Rcvd Carrier Sense Errors0 MAC Rcvd Dropped Frames 0 MAC Rcvd CRC Errors 0 MAC Rcvd Encoding Errors 0 MAC Rcvd Length Errors Large 0 MAC Rcvd Length Errors Small 0 MAC Rcvd Multicast Frames178239 MAC Rcvd Broadcast Frames48 iSCSI Port 2 Interrupt Count 182729151 Target Command Count 97555857 Initiator Command Count 0 MAC Xmit Frames 120067355 MAC Xmit Byte Count 27948414986 MAC Xmit Multicast Frames0 MAC Xmit Broadcast Frames0 MAC Xmit Pause Frames0 MAC Xmit Control Frames 0 MAC Xmit Deferrals 0 MAC Xmit Late Collisions 0 MAC Xmit Aborted 0 MAC Xmit Single Collisions 0 MAC Xmit Multiple Collisions 0 MAC Xmit Collisions 0 MAC Xmit Dropped Frames 0 MAC Xmit Jumbo Frames1718670 MAC Rcvd Frames 156126093 MAC Rcvd Byte Count 196255090216 MAC Rcvd Unknown Control Frames 0 MAC Rcvd Pause Frames0 MAC Rcvd Control Frames 0 MAC Rcvd Dribbles0 MAC Rcvd Frame Length Errors 0 MAC Rcvd Jabbers 0 MAC Rcvd Carrier Sense Errors0 MAC Rcvd Dropped Frames 0 MAC Rcvd CRC Errors 0 MAC Rcvd Encoding Errors 0 MAC Rcvd Length Errors Large 0 MAC Rcvd Length Errors Small 0