md/raid5: improve consistency of error messages.
authorNeilBrown <neilb@suse.de>
Mon, 3 May 2010 04:09:02 +0000 (14:09 +1000)
committerNeilBrown <neilb@suse.de>
Tue, 18 May 2010 05:27:58 +0000 (15:27 +1000)
Many 'printk' messages from the raid456 module mention 'raid5' even
though it may be a 'raid6' or even 'raid4' array.  This can cause
confusion.
Also the actual array name is not always reported and when it is
it is not reported consistently.

So change all the messages to start:
    md/raid:%s:
where '%s' becomes e.g. md3 to identify the particular array.

Signed-off-by: NeilBrown <neilb@suse.de>
drivers/md/raid5.c

index 81563b7c03571b2b8a629a4a8054b79d1130310f..cee9f93b35c4ca8a77124e3ed81e4f8fc21d6303 100644 (file)
@@ -1509,7 +1509,7 @@ static void raid5_end_read_request(struct bio * bi, int error)
                set_bit(R5_UPTODATE, &sh->dev[i].flags);
                if (test_bit(R5_ReadError, &sh->dev[i].flags)) {
                        rdev = conf->disks[i].rdev;
-                       printk_rl(KERN_INFO "raid5:%s: read error corrected"
+                       printk_rl(KERN_INFO "md/raid:%s: read error corrected"
                                  " (%lu sectors at %llu on %s)\n",
                                  mdname(conf->mddev), STRIPE_SECTORS,
                                  (unsigned long long)(sh->sector
@@ -1529,7 +1529,7 @@ static void raid5_end_read_request(struct bio * bi, int error)
                atomic_inc(&rdev->read_errors);
                if (conf->mddev->degraded)
                        printk_rl(KERN_WARNING
-                                 "raid5:%s: read error not correctable "
+                                 "md/raid:%s: read error not correctable "
                                  "(sector %llu on %s).\n",
                                  mdname(conf->mddev),
                                  (unsigned long long)(sh->sector
@@ -1538,7 +1538,7 @@ static void raid5_end_read_request(struct bio * bi, int error)
                else if (test_bit(R5_ReWrite, &sh->dev[i].flags))
                        /* Oh, no!!! */
                        printk_rl(KERN_WARNING
-                                 "raid5:%s: read error NOT corrected!! "
+                                 "md/raid:%s: read error NOT corrected!! "
                                  "(sector %llu on %s).\n",
                                  mdname(conf->mddev),
                                  (unsigned long long)(sh->sector
@@ -1547,7 +1547,7 @@ static void raid5_end_read_request(struct bio * bi, int error)
                else if (atomic_read(&rdev->read_errors)
                         > conf->max_nr_stripes)
                        printk(KERN_WARNING
-                              "raid5:%s: Too many read errors, failing device %s.\n",
+                              "md/raid:%s: Too many read errors, failing device %s.\n",
                               mdname(conf->mddev), bdn);
                else
                        retry = 1;
@@ -1620,7 +1620,7 @@ static void error(mddev_t *mddev, mdk_rdev_t *rdev)
 {
        char b[BDEVNAME_SIZE];
        raid5_conf_t *conf = mddev->private;
-       pr_debug("raid5: error called\n");
+       pr_debug("raid456: error called\n");
 
        if (!test_bit(Faulty, &rdev->flags)) {
                set_bit(MD_CHANGE_DEVS, &mddev->flags);
@@ -1636,9 +1636,13 @@ static void error(mddev_t *mddev, mdk_rdev_t *rdev)
                }
                set_bit(Faulty, &rdev->flags);
                printk(KERN_ALERT
-                      "raid5: Disk failure on %s, disabling device.\n"
-                      "raid5: Operation continuing on %d devices.\n",
-                      bdevname(rdev->bdev,b), conf->raid_disks - mddev->degraded);
+                      "md/raid:%s: Disk failure on %s, disabling device.\n"
+                      KERN_ALERT
+                      "md/raid:%s: Operation continuing on %d devices.\n",
+                      mdname(mddev),
+                      bdevname(rdev->bdev, b),
+                      mdname(mddev),
+                      conf->raid_disks - mddev->degraded);
        }
 }
 
@@ -1719,8 +1723,6 @@ static sector_t raid5_compute_sector(raid5_conf_t *conf, sector_t r_sector,
                        pd_idx = data_disks;
                        break;
                default:
-                       printk(KERN_ERR "raid5: unsupported algorithm %d\n",
-                               algorithm);
                        BUG();
                }
                break;
@@ -1836,10 +1838,7 @@ static sector_t raid5_compute_sector(raid5_conf_t *conf, sector_t r_sector,
                        qd_idx = raid_disks - 1;
                        break;
 
-
                default:
-                       printk(KERN_CRIT "raid6: unsupported algorithm %d\n",
-                              algorithm);
                        BUG();
                }
                break;
@@ -1902,8 +1901,6 @@ static sector_t compute_blocknr(struct stripe_head *sh, int i, int previous)
                case ALGORITHM_PARITY_N:
                        break;
                default:
-                       printk(KERN_ERR "raid5: unsupported algorithm %d\n",
-                              algorithm);
                        BUG();
                }
                break;
@@ -1962,8 +1959,6 @@ static sector_t compute_blocknr(struct stripe_head *sh, int i, int previous)
                        i -= 1;
                        break;
                default:
-                       printk(KERN_CRIT "raid6: unsupported algorithm %d\n",
-                              algorithm);
                        BUG();
                }
                break;
@@ -1976,7 +1971,8 @@ static sector_t compute_blocknr(struct stripe_head *sh, int i, int previous)
                                     previous, &dummy1, &sh2);
        if (check != sh->sector || dummy1 != dd_idx || sh2.pd_idx != sh->pd_idx
                || sh2.qd_idx != sh->qd_idx) {
-               printk(KERN_ERR "compute_blocknr: map not correct\n");
+               printk(KERN_ERR "md/raid:%s: compute_blocknr: map not correct\n",
+                      mdname(conf->mddev));
                return 0;
        }
        return r_sector;
@@ -3942,7 +3938,7 @@ static int make_request(mddev_t *mddev, struct bio * bi)
                new_sector = raid5_compute_sector(conf, logical_sector,
                                                  previous,
                                                  &dd_idx, NULL);
-               pr_debug("raid5: make_request, sector %llu logical %llu\n",
+               pr_debug("raid456: make_request, sector %llu logical %llu\n",
                        (unsigned long long)new_sector, 
                        (unsigned long long)logical_sector);
 
@@ -4721,7 +4717,7 @@ static raid5_conf_t *setup_conf(mddev_t *mddev)
        if (mddev->new_level != 5
            && mddev->new_level != 4
            && mddev->new_level != 6) {
-               printk(KERN_ERR "raid5: %s: raid level not set to 4/5/6 (%d)\n",
+               printk(KERN_ERR "md/raid:%s: raid level not set to 4/5/6 (%d)\n",
                       mdname(mddev), mddev->new_level);
                return ERR_PTR(-EIO);
        }
@@ -4729,12 +4725,12 @@ static raid5_conf_t *setup_conf(mddev_t *mddev)
             && !algorithm_valid_raid5(mddev->new_layout)) ||
            (mddev->new_level == 6
             && !algorithm_valid_raid6(mddev->new_layout))) {
-               printk(KERN_ERR "raid5: %s: layout %d not supported\n",
+               printk(KERN_ERR "md/raid:%s: layout %d not supported\n",
                       mdname(mddev), mddev->new_layout);
                return ERR_PTR(-EIO);
        }
        if (mddev->new_level == 6 && mddev->raid_disks < 4) {
-               printk(KERN_ERR "raid6: not enough configured devices for %s (%d, minimum 4)\n",
+               printk(KERN_ERR "md/raid:%s: not enough configured devices (%d, minimum 4)\n",
                       mdname(mddev), mddev->raid_disks);
                return ERR_PTR(-EINVAL);
        }
@@ -4742,8 +4738,8 @@ static raid5_conf_t *setup_conf(mddev_t *mddev)
        if (!mddev->new_chunk_sectors ||
            (mddev->new_chunk_sectors << 9) % PAGE_SIZE ||
            !is_power_of_2(mddev->new_chunk_sectors)) {
-               printk(KERN_ERR "raid5: invalid chunk size %d for %s\n",
-                      mddev->new_chunk_sectors << 9, mdname(mddev));
+               printk(KERN_ERR "md/raid:%s: invalid chunk size %d\n",
+                      mdname(mddev), mddev->new_chunk_sectors << 9);
                return ERR_PTR(-EINVAL);
        }
 
@@ -4785,7 +4781,7 @@ static raid5_conf_t *setup_conf(mddev_t *mddev)
        if (raid5_alloc_percpu(conf) != 0)
                goto abort;
 
-       pr_debug("raid5: run(%s) called.\n", mdname(mddev));
+       pr_debug("raid456: run(%s) called.\n", mdname(mddev));
 
        list_for_each_entry(rdev, &mddev->disks, same_set) {
                raid_disk = rdev->raid_disk;
@@ -4798,9 +4794,9 @@ static raid5_conf_t *setup_conf(mddev_t *mddev)
 
                if (test_bit(In_sync, &rdev->flags)) {
                        char b[BDEVNAME_SIZE];
-                       printk(KERN_INFO "raid5: device %s operational as raid"
-                               " disk %d\n", bdevname(rdev->bdev,b),
-                               raid_disk);
+                       printk(KERN_INFO "md/raid:%s: device %s operational as raid"
+                              " disk %d\n",
+                              mdname(mddev), bdevname(rdev->bdev, b), raid_disk);
                } else
                        /* Cannot rely on bitmap to complete recovery */
                        conf->fullsync = 1;
@@ -4824,16 +4820,17 @@ static raid5_conf_t *setup_conf(mddev_t *mddev)
                 max_disks * ((sizeof(struct bio) + PAGE_SIZE))) / 1024;
        if (grow_stripes(conf, conf->max_nr_stripes)) {
                printk(KERN_ERR
-                       "raid5: couldn't allocate %dkB for buffers\n", memory);
+                      "md/raid:%s: couldn't allocate %dkB for buffers\n",
+                      mdname(mddev), memory);
                goto abort;
        } else
-               printk(KERN_INFO "raid5: allocated %dkB for %s\n",
-                       memory, mdname(mddev));
+               printk(KERN_INFO "md/raid:%s: allocated %dkB\n",
+                      mdname(mddev), memory);
 
        conf->thread = md_register_thread(raid5d, mddev, NULL);
        if (!conf->thread) {
                printk(KERN_ERR
-                      "raid5: couldn't allocate thread for %s\n",
+                      "md/raid:%s: couldn't allocate thread.\n",
                       mdname(mddev));
                goto abort;
        }
@@ -4884,7 +4881,7 @@ static int run(mddev_t *mddev)
        sector_t reshape_offset = 0;
 
        if (mddev->recovery_cp != MaxSector)
-               printk(KERN_NOTICE "raid5: %s is not clean"
+               printk(KERN_NOTICE "md/raid:%s: not clean"
                       " -- starting background reconstruction\n",
                       mdname(mddev));
        if (mddev->reshape_position != MaxSector) {
@@ -4898,7 +4895,7 @@ static int run(mddev_t *mddev)
                int max_degraded = (mddev->level == 6 ? 2 : 1);
 
                if (mddev->new_level != mddev->level) {
-                       printk(KERN_ERR "raid5: %s: unsupported reshape "
+                       printk(KERN_ERR "md/raid:%s: unsupported reshape "
                               "required - aborting.\n",
                               mdname(mddev));
                        return -EINVAL;
@@ -4911,8 +4908,8 @@ static int run(mddev_t *mddev)
                here_new = mddev->reshape_position;
                if (sector_div(here_new, mddev->new_chunk_sectors *
                               (mddev->raid_disks - max_degraded))) {
-                       printk(KERN_ERR "raid5: reshape_position not "
-                              "on a stripe boundary\n");
+                       printk(KERN_ERR "md/raid:%s: reshape_position not "
+                              "on a stripe boundary\n", mdname(mddev));
                        return -EINVAL;
                }
                reshape_offset = here_new * mddev->new_chunk_sectors;
@@ -4933,8 +4930,9 @@ static int run(mddev_t *mddev)
                        if ((here_new * mddev->new_chunk_sectors != 
                             here_old * mddev->chunk_sectors) ||
                            mddev->ro == 0) {
-                               printk(KERN_ERR "raid5: in-place reshape must be started"
-                                      " in read-only mode - aborting\n");
+                               printk(KERN_ERR "md/raid:%s: in-place reshape must be started"
+                                      " in read-only mode - aborting\n",
+                                      mdname(mddev));
                                return -EINVAL;
                        }
                } else if (mddev->delta_disks < 0
@@ -4943,11 +4941,13 @@ static int run(mddev_t *mddev)
                    : (here_new * mddev->new_chunk_sectors >=
                       here_old * mddev->chunk_sectors)) {
                        /* Reading from the same stripe as writing to - bad */
-                       printk(KERN_ERR "raid5: reshape_position too early for "
-                              "auto-recovery - aborting.\n");
+                       printk(KERN_ERR "md/raid:%s: reshape_position too early for "
+                              "auto-recovery - aborting.\n",
+                              mdname(mddev));
                        return -EINVAL;
                }
-               printk(KERN_INFO "raid5: reshape will continue\n");
+               printk(KERN_INFO "md/raid:%s: reshape will continue\n",
+                      mdname(mddev));
                /* OK, we should be able to continue; */
        } else {
                BUG_ON(mddev->level != mddev->new_level);
@@ -4989,18 +4989,6 @@ static int run(mddev_t *mddev)
                    mddev->minor_version > 90)
                        rdev->recovery_offset = reshape_offset;
                        
-               printk("%d: w=%d pa=%d pr=%d m=%d a=%d r=%d op1=%d op2=%d\n",
-                      rdev->raid_disk, working_disks, conf->prev_algo,
-                      conf->previous_raid_disks, conf->max_degraded,
-                      conf->algorithm, conf->raid_disks, 
-                      only_parity(rdev->raid_disk,
-                                  conf->prev_algo,
-                                  conf->previous_raid_disks,
-                                  conf->max_degraded),
-                      only_parity(rdev->raid_disk,
-                                  conf->algorithm,
-                                  conf->raid_disks,
-                                  conf->max_degraded));
                if (rdev->recovery_offset < reshape_offset) {
                        /* We need to check old and new layout */
                        if (!only_parity(rdev->raid_disk,
@@ -5021,7 +5009,7 @@ static int run(mddev_t *mddev)
                           - working_disks);
 
        if (mddev->degraded > conf->max_degraded) {
-               printk(KERN_ERR "raid5: not enough operational devices for %s"
+               printk(KERN_ERR "md/raid:%s: not enough operational devices"
                        " (%d/%d failed)\n",
                        mdname(mddev), mddev->degraded, conf->raid_disks);
                goto abort;
@@ -5035,32 +5023,32 @@ static int run(mddev_t *mddev)
            mddev->recovery_cp != MaxSector) {
                if (mddev->ok_start_degraded)
                        printk(KERN_WARNING
-                              "raid5: starting dirty degraded array: %s"
-                              "- data corruption possible.\n",
+                              "md/raid:%s: starting dirty degraded array"
+                              " - data corruption possible.\n",
                               mdname(mddev));
                else {
                        printk(KERN_ERR
-                              "raid5: cannot start dirty degraded array for %s\n",
+                              "md/raid:%s: cannot start dirty degraded array.\n",
                               mdname(mddev));
                        goto abort;
                }
        }
 
        if (mddev->degraded == 0)
-               printk("raid5: raid level %d set %s active with %d out of %d"
-                      " devices, algorithm %d\n", conf->level, mdname(mddev),
+               printk(KERN_INFO "md/raid:%s: raid level %d active with %d out of %d"
+                      " devices, algorithm %d\n", mdname(mddev), conf->level,
                       mddev->raid_disks-mddev->degraded, mddev->raid_disks,
                       mddev->new_layout);
        else
-               printk(KERN_ALERT "raid5: raid level %d set %s active with %d"
-                       " out of %d devices, algorithm %d\n", conf->level,
-                       mdname(mddev), mddev->raid_disks - mddev->degraded,
-                       mddev->raid_disks, mddev->new_layout);
+               printk(KERN_ALERT "md/raid:%s: raid level %d active with %d"
+                      " out of %d devices, algorithm %d\n",
+                      mdname(mddev), conf->level,
+                      mddev->raid_disks - mddev->degraded,
+                      mddev->raid_disks, mddev->new_layout);
 
        print_raid5_conf(conf);
 
        if (conf->reshape_progress != MaxSector) {
-               printk("...ok start reshape thread\n");
                conf->reshape_safe = conf->reshape_progress;
                atomic_set(&conf->reshape_stripes, 0);
                clear_bit(MD_RECOVERY_SYNC, &mddev->recovery);
@@ -5087,7 +5075,7 @@ static int run(mddev_t *mddev)
                mddev->to_remove = NULL;
        else if (sysfs_create_group(&mddev->kobj, &raid5_attrs_group))
                printk(KERN_WARNING
-                      "raid5: failed to create sysfs attributes for %s\n",
+                      "md/raid:%s: failed to create sysfs attributes.\n",
                       mdname(mddev));
 
        mddev->queue->queue_lock = &conf->device_lock;
@@ -5117,12 +5105,10 @@ abort:
                free_conf(conf);
        }
        mddev->private = NULL;
-       printk(KERN_ALERT "raid5: failed to run raid set %s\n", mdname(mddev));
+       printk(KERN_ALERT "md/raid:%s: failed to run raid set.\n", mdname(mddev));
        return -EIO;
 }
 
-
-
 static int stop(mddev_t *mddev)
 {
        raid5_conf_t *conf = mddev->private;
@@ -5196,21 +5182,22 @@ static void print_raid5_conf (raid5_conf_t *conf)
        int i;
        struct disk_info *tmp;
 
-       printk("RAID5 conf printout:\n");
+       printk(KERN_DEBUG "RAID conf printout:\n");
        if (!conf) {
                printk("(conf==NULL)\n");
                return;
        }
-       printk(" --- rd:%d wd:%d\n", conf->raid_disks,
-                conf->raid_disks - conf->mddev->degraded);
+       printk(KERN_DEBUG " --- level:%d rd:%d wd:%d\n", conf->level,
+              conf->raid_disks,
+              conf->raid_disks - conf->mddev->degraded);
 
        for (i = 0; i < conf->raid_disks; i++) {
                char b[BDEVNAME_SIZE];
                tmp = conf->disks + i;
                if (tmp->rdev)
-               printk(" disk %d, o:%d, dev:%s\n",
-                       i, !test_bit(Faulty, &tmp->rdev->flags),
-                       bdevname(tmp->rdev->bdev,b));
+                       printk(KERN_DEBUG " disk %d, o:%d, dev:%s\n",
+                              i, !test_bit(Faulty, &tmp->rdev->flags),
+                              bdevname(tmp->rdev->bdev, b));
        }
 }
 
@@ -5358,7 +5345,8 @@ static int check_stripe_cache(mddev_t *mddev)
            > conf->max_nr_stripes ||
            ((mddev->new_chunk_sectors << 9) / STRIPE_SIZE) * 4
            > conf->max_nr_stripes) {
-               printk(KERN_WARNING "raid5: reshape: not enough stripes.  Needed %lu\n",
+               printk(KERN_WARNING "md/raid:%s: reshape: not enough stripes.  Needed %lu\n",
+                      mdname(mddev),
                       ((max(mddev->chunk_sectors, mddev->new_chunk_sectors) << 9)
                        / STRIPE_SIZE)*4);
                return 0;
@@ -5429,7 +5417,7 @@ static int raid5_start_reshape(mddev_t *mddev)
         */
        if (raid5_size(mddev, 0, conf->raid_disks + mddev->delta_disks)
            < mddev->array_sectors) {
-               printk(KERN_ERR "md%s: array size must be reduced "
+               printk(KERN_ERR "md/raid:%s: array size must be reduced "
                       "before number of disks\n", mdname(mddev));
                return -EINVAL;
        }
@@ -5467,9 +5455,9 @@ static int raid5_start_reshape(mddev_t *mddev)
                                if (sysfs_create_link(&mddev->kobj,
                                                      &rdev->kobj, nm))
                                        printk(KERN_WARNING
-                                              "raid5: failed to create "
-                                              " link %s for %s\n",
-                                              nm, mdname(mddev));
+                                              "md/raid:%s: failed to create "
+                                              " link %s\n",
+                                              mdname(mddev), nm);
                        } else
                                break;
                }
@@ -5616,7 +5604,8 @@ static void *raid45_takeover_raid0(mddev_t *mddev, int level)
 
        /* for raid0 takeover only one zone is supported */
        if (raid0_priv->nr_strip_zones > 1) {
-               printk(KERN_ERR "md: cannot takeover raid0 with more than one zone.\n");
+               printk(KERN_ERR "md/raid:%s: cannot takeover raid0 with more than one zone.\n",
+                      mdname(mddev));
                return ERR_PTR(-EINVAL);
        }