Re: INFO: HARDIRQ-safe - HARDIRQ-unsafe lock order detected
On Fri, Aug 10, 2012 at 2:58 AM, Michael Christie micha...@cs.wisc.edu wrote: On Aug 8, 2012, at 4:42 AM, Fubo Chen fubo.c...@gmail.com wrote: Anyone seen this before ? Also occurs with 3.4.1. == [ INFO: HARDIRQ-safe - HARDIRQ-unsafe lock order detected ] 3.6.0-rc1-debug+ #1 Not tainted -- swapper/1/0 [HC0[0]:SC1[1]:HE0:SE0] is trying to acquire: ((session-lock)-rlock){+.-...}, at: [a025dc08] iscsi_eh_cmd_timed_out+0x58/0x2e0 [libiscsi] and this task is already holding: ((q-__queue_lock)-rlock){-.-...}, at: [811f6965] blk_rq_timed_out_timer+0x25/0x140 which would create a new lock dependency: ((q-__queue_lock)-rlock){-.-...} - ((session-lock)-rlock){+.-...} but this new dependency connects a HARDIRQ-irq-safe lock: ((q-__queue_lock)-rlock){-.-...} ... which became HARDIRQ-irq-safe at: [8109b5ca] __lock_acquire+0x7ea/0x1ba0 [8109cfc2] lock_acquire+0x92/0x140 [814b41c5] _raw_spin_lock_irqsave+0x65/0xb0 [812e2974] blk_done+0x34/0x110 [81295889] vring_interrupt+0x49/0xc0 [810c68f5] handle_irq_event_percpu+0x75/0x270 [810c6b38] handle_irq_event+0x48/0x70 [810c9477] handle_edge_irq+0x77/0x110 [81004042] handle_irq+0x22/0x40 [814bda2a] do_IRQ+0x5a/0xe0 [814b436f] ret_from_intr+0x0/0x1a [8100a7da] default_idle+0x4a/0x170 [8100b609] cpu_idle+0xe9/0x130 [814a4c6e] start_secondary+0x26a/0x26c Does this error only occur when using some sort of virt setup? I do not think we will hit this with iscsi, because we do not ever grab the queue lock for a iscsi device from hard irq context. It is always done from softirq or thread context. The snippet above seems to be from the virtio_blk.c code. Yes. This happened inside KVM machine. Fubo. -- 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?hl=en.
Re: iscsid: Kernel reported iSCSI connection 1:0 error (1011) state (3)
oki Mike. On Thu, Aug 9, 2012 at 10:54 PM, Michael Christie micha...@cs.wisc.eduwrote: In the future can you post these questions to the open-iscsi mailing list? This is the last time I will respond privately. It is best ti post on the list so others can see the answer (people searching in google for similar problems in the future). It changes name because you probably still have some app/user with a reference to the device. When yourelogin the scsi layer will then give you a different name. You should not be using the sdX types of names, because they are not persistent. You should use the udev names in /dev/disk/. On Aug 9, 2012, at 3:17 AM, parveen kumar coolbudy1...@gmail.com wrote: Hi Mike Christie, ok , but what about /dev/sdb, first time its showing me /dev/sda then connection bracked and i had tried again and next its showing /dev/sdb means if network disruption again i have to edit /etc/fstab ? Regards, PARVEEN On Thu, Aug 9, 2012 at 11:26 AM, Michael Christie micha...@cs.wisc.edu wrote: There was a probably a network disruption that caused the initiator to lose the connection On Aug 9, 2012, at 12:11 AM, parveen kumar coolbudy1...@gmail.com wrote: Hello Mike Christie, Few days back i mounted 2TB volume through ISCSI from SAN box in my server. After mounting the 2TB space i am bale to see partition /dev/sda of 2TB on my server. But after 5-6 days m not able to write data on /dev/sda, then i see conncetion is lost and iscsi connection is braked, so i tried again to login to target with this command: # iscsiadm -m node -T target_iqn_name -p ipaddress -l after this, login is sucessfull, then i put command: # fdisk -l its showing me /dev/sdb of 2TB size i m not able to understand what happend, first time its showing me /dev/sda then connection bracked and i had tried again and next its showing /dev/sdb please help me to sort out this problem.. -- . -- . -- *.* -- 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?hl=en.
DataDigest in Open-iSCSI
Hi everyone, I have been setting up Open-iSCSI Initiator in order to measure its throughput and find out its bottlenecks for my research. I was trying to enable DataDigest on the Initiator but failed. I'm not sure if DataDigests are implemented in Open-iSCSI or not? I searched documentation and this group, but didn't find any reference. However, HeaderDigest are enabled and working properly, but I need DataDigest too. If Open-iSCSI doesn't support DataDigest, can you recommend me some other Linux/Open Source-based iSCSI Initiator? Grealty appreciate your help, J -- You received this message because you are subscribed to the Google Groups open-iscsi group. To view this discussion on the web visit https://groups.google.com/d/msg/open-iscsi/-/9akyK1JSJKwJ. 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?hl=en.
Re: DataDigest in Open-iSCSI
On 08/10/2012 03:08 AM, js19...@gmail.com wrote: Hi everyone, I have been setting up Open-iSCSI Initiator in order to measure its throughput and find out its bottlenecks for my research. I was trying to One of the major bottlenecks is the memcpy that is done in the recv path. See the function libiscsi_tcp.c:iscsi_tcp_segment_recv(). enable DataDigest on the Initiator but failed. I'm not sure if DataDigests are implemented in Open-iSCSI or not? I searched They are but for many uses they are not safe, because of issues in higher up levels we cannot control, so they are not enabled in many distros. documentation and this group, but didn't find any reference. However, HeaderDigest are enabled and working properly, but I need DataDigest too. Are you using RHEL/Centos? Try using the open-iscsi.org tarball version or some other distro. When you use data digests we will use memcpy instead of doing zero copy sends for data, so that will be major perf issue. -- 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?hl=en.
Re: iscsid: Kernel reported iSCSI connection 1:0 error (1011) state (3)
On 08/10/2012 04:34 AM, parveen kumar wrote: oki Mike. Thanks. One thing that is not clear to me, is why you have to do the logout then relogin? Are you doing this because you have a app or FS using the iscsi disk, and it gets IO errors when you get the conn errors? Or, are you doing the relogin because the /dev/sdX is automatically removed when you see the conn errors messages? If this is happening what version of open-iscsi are you using and where did you get it (open-iscsi.org or some distro)? In /var/log/messages do you see session recovery timed out after 120 secs (the value for the secs may be different for you if you modified it in iscsid.conf or with iscsiadm). On Thu, Aug 9, 2012 at 10:54 PM, Michael Christie micha...@cs.wisc.edu mailto:micha...@cs.wisc.edu wrote: In the future can you post these questions to the open-iscsi mailing list? This is the last time I will respond privately. It is best ti post on the list so others can see the answer (people searching in google for similar problems in the future). It changes name because you probably still have some app/user with a reference to the device. When yourelogin the scsi layer will then give you a different name. You should not be using the sdX types of names, because they are not persistent. You should use the udev names in /dev/disk/. On Aug 9, 2012, at 3:17 AM, parveen kumar coolbudy1...@gmail.com mailto:coolbudy1...@gmail.com wrote: Hi Mike Christie, ok , but what about /dev/sdb, first time its showing me /dev/sda then connection bracked and i had tried again and next its showing /dev/sdb means if network disruption again i have to edit /etc/fstab ? Regards, PARVEEN On Thu, Aug 9, 2012 at 11:26 AM, Michael Christie micha...@cs.wisc.edu mailto:micha...@cs.wisc.edu wrote: There was a probably a network disruption that caused the initiator to lose the connection On Aug 9, 2012, at 12:11 AM, parveen kumar coolbudy1...@gmail.com mailto:coolbudy1...@gmail.com wrote: Hello Mike Christie, Few days back i mounted 2TB volume through ISCSI from SAN box in my server. After mounting the 2TB space i am bale to see partition /dev/sda of 2TB on my server. But after 5-6 days m not able to write data on /dev/sda, then i see conncetion is lost and iscsi connection is braked, so i tried again to login to target with this command: # iscsiadm -m node -T target_iqn_name -p ipaddress -l after this, login is sucessfull, then i put command: # fdisk -l its showing me /dev/sdb of 2TB size i m not able to understand what happend, first time its showing me /dev/sda then connection bracked and i had tried again and next its showing /dev/sdb please help me to sort out this problem.. -- . -- . -- *.* -- 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?hl=en. -- 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?hl=en.