* [PATCH md ] Fix up handling for read error in raid1. [not found] <20040624143619.5067.patches@notabene> @ 2004-06-24 4:37 ` NeilBrown 2004-07-12 14:29 ` Maurilio Longo 0 siblings, 1 reply; 4+ messages in thread From: NeilBrown @ 2004-06-24 4:37 UTC (permalink / raw) To: Andrew Morton; +Cc: linux-raid There is severe bit-rot in this code, which is to say that it doesn't work at all: an io error during read will do bad things. It should work better with this patch. Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au> ### Diffstat output ./drivers/md/raid1.c | 10 +++++++--- 1 files changed, 7 insertions(+), 3 deletions(-) diff ./drivers/md/raid1.c~current~ ./drivers/md/raid1.c --- ./drivers/md/raid1.c~current~ 2004-06-23 14:37:05.000000000 +1000 +++ ./drivers/md/raid1.c 2004-06-23 14:37:13.000000000 +1000 @@ -206,7 +206,7 @@ static int map(mddev_t *mddev, mdk_rdev_ *rdevp = rdev; atomic_inc(&rdev->nr_pending); spin_unlock_irq(&conf->device_lock); - return 0; + return i; } } spin_unlock_irq(&conf->device_lock); @@ -945,18 +945,22 @@ static void raid1d(mddev_t *mddev) mddev = r1_bio->mddev; conf = mddev_to_conf(mddev); - bio = r1_bio->master_bio; if (test_bit(R1BIO_IsSync, &r1_bio->state)) { sync_request_write(mddev, r1_bio); unplug = 1; } else { - if (map(mddev, &rdev) == -1) { + int disk; + bio = r1_bio->bios[r1_bio->read_disk]; + if ((disk=map(mddev, &rdev)) == -1) { printk(KERN_ALERT "raid1: %s: unrecoverable I/O" " read error for block %llu\n", bdevname(bio->bi_bdev,b), (unsigned long long)r1_bio->sector); raid_end_bio_io(r1_bio); } else { + r1_bio->bios[r1_bio->read_disk] = NULL; + r1_bio->read_disk = disk; + r1_bio->bios[r1_bio->read_disk] = bio; printk(KERN_ERR "raid1: %s: redirecting sector %llu to" " another mirror\n", bdevname(rdev->bdev,b), ^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH md ] Fix up handling for read error in raid1. 2004-06-24 4:37 ` [PATCH md ] Fix up handling for read error in raid1 NeilBrown @ 2004-07-12 14:29 ` Maurilio Longo 2004-07-23 2:56 ` Neil Brown 0 siblings, 1 reply; 4+ messages in thread From: Maurilio Longo @ 2004-07-12 14:29 UTC (permalink / raw) To: linux raid mailing list; +Cc: NeilBrown Neil, do you have a similar patch for a 2.4.x (x == 20 in my case) kernel? I can't change kernel easily since I use it to run mars-nwe which is a novell emulator which is going to be removed from every distro I'm aware of :( regards. NeilBrown ha scritto: > There is severe bit-rot in this code, which is to say that it doesn't work > at all: an io error during read will do bad things. It should work > better with this patch. > > Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au> > > ### Diffstat output > ./drivers/md/raid1.c | 10 +++++++--- > 1 files changed, 7 insertions(+), 3 deletions(-) > > diff ./drivers/md/raid1.c~current~ ./drivers/md/raid1.c > --- ./drivers/md/raid1.c~current~ 2004-06-23 14:37:05.000000000 +1000 > +++ ./drivers/md/raid1.c 2004-06-23 14:37:13.000000000 +1000 > @@ -206,7 +206,7 @@ static int map(mddev_t *mddev, mdk_rdev_ > *rdevp = rdev; > atomic_inc(&rdev->nr_pending); > spin_unlock_irq(&conf->device_lock); > - return 0; > + return i; > } > } > spin_unlock_irq(&conf->device_lock); > @@ -945,18 +945,22 @@ static void raid1d(mddev_t *mddev) > > mddev = r1_bio->mddev; > conf = mddev_to_conf(mddev); > - bio = r1_bio->master_bio; > if (test_bit(R1BIO_IsSync, &r1_bio->state)) { > sync_request_write(mddev, r1_bio); > unplug = 1; > } else { > - if (map(mddev, &rdev) == -1) { > + int disk; > + bio = r1_bio->bios[r1_bio->read_disk]; > + if ((disk=map(mddev, &rdev)) == -1) { > printk(KERN_ALERT "raid1: %s: unrecoverable I/O" > " read error for block %llu\n", > bdevname(bio->bi_bdev,b), > (unsigned long long)r1_bio->sector); > raid_end_bio_io(r1_bio); > } else { > + r1_bio->bios[r1_bio->read_disk] = NULL; > + r1_bio->read_disk = disk; > + r1_bio->bios[r1_bio->read_disk] = bio; > printk(KERN_ERR "raid1: %s: redirecting sector %llu to" > " another mirror\n", > bdevname(rdev->bdev,b), > - > To unsubscribe from this list: send the line "unsubscribe linux-raid" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html -- __________ | | | |__| md2520@mclink.it |_|_|_|____| ^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH md ] Fix up handling for read error in raid1. 2004-07-12 14:29 ` Maurilio Longo @ 2004-07-23 2:56 ` Neil Brown 2004-07-23 7:26 ` Maurilio Longo 0 siblings, 1 reply; 4+ messages in thread From: Neil Brown @ 2004-07-23 2:56 UTC (permalink / raw) To: Maurilio Longo; +Cc: linux raid mailing list On Monday July 12, maurilio.longo@libero.it wrote: > Neil, > > do you have a similar patch for a 2.4.x (x == 20 in my case) kernel? > > I can't change kernel easily since I use it to run mars-nwe which is a novell emulator > which is going to be removed from every distro I'm aware of :( This bug is specific to 2.6. There is no need for anything similar in 2.4 NeilBrown > > regards. > > > NeilBrown ha scritto: > > > There is severe bit-rot in this code, which is to say that it doesn't work > > at all: an io error during read will do bad things. It should work > > better with this patch. > > > > Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au> > > > > ### Diffstat output > > ./drivers/md/raid1.c | 10 +++++++--- > > 1 files changed, 7 insertions(+), 3 deletions(-) > > > > diff ./drivers/md/raid1.c~current~ ./drivers/md/raid1.c > > --- ./drivers/md/raid1.c~current~ 2004-06-23 14:37:05.000000000 +1000 > > +++ ./drivers/md/raid1.c 2004-06-23 14:37:13.000000000 +1000 > > @@ -206,7 +206,7 @@ static int map(mddev_t *mddev, mdk_rdev_ > > *rdevp = rdev; > > atomic_inc(&rdev->nr_pending); > > spin_unlock_irq(&conf->device_lock); > > - return 0; > > + return i; > > } > > } > > spin_unlock_irq(&conf->device_lock); > > @@ -945,18 +945,22 @@ static void raid1d(mddev_t *mddev) > > > > mddev = r1_bio->mddev; > > conf = mddev_to_conf(mddev); > > - bio = r1_bio->master_bio; > > if (test_bit(R1BIO_IsSync, &r1_bio->state)) { > > sync_request_write(mddev, r1_bio); > > unplug = 1; > > } else { > > - if (map(mddev, &rdev) == -1) { > > + int disk; > > + bio = r1_bio->bios[r1_bio->read_disk]; > > + if ((disk=map(mddev, &rdev)) == -1) { > > printk(KERN_ALERT "raid1: %s: unrecoverable I/O" > > " read error for block %llu\n", > > bdevname(bio->bi_bdev,b), > > (unsigned long long)r1_bio->sector); > > raid_end_bio_io(r1_bio); > > } else { > > + r1_bio->bios[r1_bio->read_disk] = NULL; > > + r1_bio->read_disk = disk; > > + r1_bio->bios[r1_bio->read_disk] = bio; > > printk(KERN_ERR "raid1: %s: redirecting sector %llu to" > > " another mirror\n", > > bdevname(rdev->bdev,b), > > - > > To unsubscribe from this list: send the line "unsubscribe linux-raid" in > > the body of a message to majordomo@vger.kernel.org > > More majordomo info at http://vger.kernel.org/majordomo-info.html > > -- > __________ > | | | |__| md2520@mclink.it > |_|_|_|____| > ^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH md ] Fix up handling for read error in raid1. 2004-07-23 2:56 ` Neil Brown @ 2004-07-23 7:26 ` Maurilio Longo 0 siblings, 0 replies; 4+ messages in thread From: Maurilio Longo @ 2004-07-23 7:26 UTC (permalink / raw) To: Neil Brown; +Cc: linux raid mailing list Neil, I was talking about a read error on a raid1 disk which causes that disk to be removed from array before trying to do a read on the other disk followed by a write to the failed one to force disk to remap broken sector. It seems to me 2.4.x raid1.c code doesn't has got this feature. regards. ps. probably I simply don't understand what raid1.c code is doing in 2.4.x and 2.6.x Neil Brown ha scritto: > On Monday July 12, maurilio.longo@libero.it wrote: > > Neil, > > > > do you have a similar patch for a 2.4.x (x == 20 in my case) kernel? > > > > I can't change kernel easily since I use it to run mars-nwe which is a novell emulator > > which is going to be removed from every distro I'm aware of :( > > This bug is specific to 2.6. There is no need for anything similar in > 2.4 > > NeilBrown > > > > > regards. > > > > > > NeilBrown ha scritto: > > > > > There is severe bit-rot in this code, which is to say that it doesn't work > > > at all: an io error during read will do bad things. It should work > > > better with this patch. > > > > > > Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au> > > > > > > ### Diffstat output > > > ./drivers/md/raid1.c | 10 +++++++--- > > > 1 files changed, 7 insertions(+), 3 deletions(-) > > > > > > diff ./drivers/md/raid1.c~current~ ./drivers/md/raid1.c > > > --- ./drivers/md/raid1.c~current~ 2004-06-23 14:37:05.000000000 +1000 > > > +++ ./drivers/md/raid1.c 2004-06-23 14:37:13.000000000 +1000 > > > @@ -206,7 +206,7 @@ static int map(mddev_t *mddev, mdk_rdev_ > > > *rdevp = rdev; > > > atomic_inc(&rdev->nr_pending); > > > spin_unlock_irq(&conf->device_lock); > > > - return 0; > > > + return i; > > > } > > > } > > > spin_unlock_irq(&conf->device_lock); > > > @@ -945,18 +945,22 @@ static void raid1d(mddev_t *mddev) > > > > > > mddev = r1_bio->mddev; > > > conf = mddev_to_conf(mddev); > > > - bio = r1_bio->master_bio; > > > if (test_bit(R1BIO_IsSync, &r1_bio->state)) { > > > sync_request_write(mddev, r1_bio); > > > unplug = 1; > > > } else { > > > - if (map(mddev, &rdev) == -1) { > > > + int disk; > > > + bio = r1_bio->bios[r1_bio->read_disk]; > > > + if ((disk=map(mddev, &rdev)) == -1) { > > > printk(KERN_ALERT "raid1: %s: unrecoverable I/O" > > > " read error for block %llu\n", > > > bdevname(bio->bi_bdev,b), > > > (unsigned long long)r1_bio->sector); > > > raid_end_bio_io(r1_bio); > > > } else { > > > + r1_bio->bios[r1_bio->read_disk] = NULL; > > > + r1_bio->read_disk = disk; > > > + r1_bio->bios[r1_bio->read_disk] = bio; > > > printk(KERN_ERR "raid1: %s: redirecting sector %llu to" > > > " another mirror\n", > > > bdevname(rdev->bdev,b), > > > - > > > To unsubscribe from this list: send the line "unsubscribe linux-raid" in > > > the body of a message to majordomo@vger.kernel.org > > > More majordomo info at http://vger.kernel.org/majordomo-info.html > > > > -- > > __________ > > | | | |__| md2520@mclink.it > > |_|_|_|____| > > > - > To unsubscribe from this list: send the line "unsubscribe linux-raid" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html -- __________ | | | |__| md2520@mclink.it |_|_|_|____| Team OS/2 Italia ^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2004-07-23 7:26 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <20040624143619.5067.patches@notabene>
2004-06-24 4:37 ` [PATCH md ] Fix up handling for read error in raid1 NeilBrown
2004-07-12 14:29 ` Maurilio Longo
2004-07-23 2:56 ` Neil Brown
2004-07-23 7:26 ` Maurilio Longo
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).