Re: MDADM 3.3 broken?
From: NeilBrown <hidden>
Date: 2014-01-20 04:34:21
Subsystem:
the rest · Maintainer:
Linus Torvalds
On Sat, 14 Dec 2013 13:01:50 -0800 "David F." [off-list ref] wrote:
Hi, Just wondering if this gave you guys everything you needed to figure out the issue?
I had everything but time. I've now made the time and have the fix (I hope). Please try the current HEAD of git://neil.brown.name/mdadm/ The important patch is below.
Also, any idea on when 3.4 may be out with the various fixes?
I hope to release 3.3.1 some time in February. Based on past experience it should be out before Easter, but no promises. NeilBrown From f0e876ce03a63f150bb87b2734c139bc8bb285b2 Mon Sep 17 00:00:00 2001 From: NeilBrown <redacted> Date: Mon, 20 Jan 2014 15:27:29 +1100 Subject: [PATCH] DDF: fix detection of failed devices during assembly. When we call "getinfo_super", we report the working/failed status of the particular device, and also (via the 'map') the working/failed status of every other device that this metadata is aware of. It is important that the way we calculate "working or failed" is consistent. As it is, getinfo_super_ddf() will report a spare as "working", but every other device will see it as "failed", which leads to failure to assemble arrays with spares. For getinfo_super_ddf (i.e. for the container), a device is assumed "working" unless flagged as DDF_Failed. For getinfo_super_ddf_bvd (for a member array), a device is assumed "failed" unless DDF_Online is set, and DDF_Failed is not set. Reported-by: "David F." <redacted> Signed-off-by: NeilBrown <redacted>
diff --git a/super-ddf.c b/super-ddf.c
index d526d8ad3da9..4242af86fea9 100644
--- a/super-ddf.c
+++ b/super-ddf.c@@ -1913,6 +1913,7 @@ static void getinfo_super_ddf(struct supertype *st, struct mdinfo *info, char *m info->disk.major = 0; info->disk.minor = 0; if (ddf->dlist) { + struct phys_disk_entry *pde = NULL; info->disk.number = be32_to_cpu(ddf->dlist->disk.refnum); info->disk.raid_disk = find_phys(ddf, ddf->dlist->disk.refnum);
@@ -1920,12 +1921,19 @@ static void getinfo_super_ddf(struct supertype *st, struct mdinfo *info, char *m entries[info->disk.raid_disk]. config_size); info->component_size = ddf->dlist->size - info->data_offset; + if (info->disk.raid_disk >= 0) + pde = ddf->phys->entries + info->disk.raid_disk; + if (pde && + !(be16_to_cpu(pde->state) & DDF_Failed)) + info->disk.state = (1 << MD_DISK_SYNC) | (1 << MD_DISK_ACTIVE); + else + info->disk.state = 1 << MD_DISK_FAULTY; } else { info->disk.number = -1; info->disk.raid_disk = -1; // info->disk.raid_disk = find refnum in the table and use index; + info->disk.state = (1 << MD_DISK_SYNC) | (1 << MD_DISK_ACTIVE); } - info->disk.state = (1 << MD_DISK_SYNC) | (1 << MD_DISK_ACTIVE); info->recovery_start = MaxSector; info->reshape_active = 0;
@@ -1943,8 +1951,6 @@ static void getinfo_super_ddf(struct supertype *st, struct mdinfo *info, char *m int i; for (i = 0 ; i < map_disks; i++) { if (i < info->array.raid_disks && - (be16_to_cpu(ddf->phys->entries[i].state) - & DDF_Online) && !(be16_to_cpu(ddf->phys->entries[i].state) & DDF_Failed)) map[i] = 1;
@@ -2017,7 +2023,11 @@ static void getinfo_super_ddf_bvd(struct supertype *st, struct mdinfo *info, cha info->disk.raid_disk = cd + conf->sec_elmnt_seq * be16_to_cpu(conf->prim_elmnt_count); info->disk.number = dl->pdnum; - info->disk.state = (1<<MD_DISK_SYNC)|(1<<MD_DISK_ACTIVE); + info->disk.state = 0; + if (info->disk.number >= 0 && + (be16_to_cpu(ddf->phys->entries[info->disk.number].state) & DDF_Online) && + !(be16_to_cpu(ddf->phys->entries[info->disk.number].state) & DDF_Failed)) + info->disk.state = (1<<MD_DISK_SYNC)|(1<<MD_DISK_ACTIVE); } info->container_member = ddf->currentconf->vcnum;
Attachments
- signature.asc [application/pgp-signature] 828 bytes