From: Bob Peterson <rpeterso@redhat.com>
To: cluster-devel.redhat.com
Subject: [Cluster-devel] [GFS2 PATCH 2/5] gfs2: instrumentation wrt ail1 stuck
Date: Thu, 26 Mar 2020 13:40:17 -0500 [thread overview]
Message-ID: <20200326184020.123544-3-rpeterso@redhat.com> (raw)
In-Reply-To: <20200326184020.123544-1-rpeterso@redhat.com>
Before this patch, if the ail1 flush got stuck for some reason, there
were no clues as to why. This patch introduces a check for getting
stuck for more than a minute, and if it happens, it dumps the items
still remaining on the ail1 list.
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
---
fs/gfs2/log.c | 41 +++++++++++++++++++++++++++++++++++++++++
1 file changed, 41 insertions(+)
diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c
index 87f3e892be3e..2abec43ae898 100644
--- a/fs/gfs2/log.c
+++ b/fs/gfs2/log.c
@@ -139,6 +139,41 @@ __acquires(&sdp->sd_ail_lock)
return ret;
}
+static void dump_ail_list(struct gfs2_sbd *sdp)
+{
+ struct gfs2_trans *tr;
+ struct gfs2_bufdata *bd;
+ struct buffer_head *bh;
+
+ fs_err(sdp, "Error: In gfs2_ail1_flush for a minute! t=%d\n",
+ current->journal_info ? 1 : 0);
+
+ list_for_each_entry_reverse(tr, &sdp->sd_ail1_list, tr_list) {
+ list_for_each_entry_reverse(bd, &tr->tr_ail1_list,
+ bd_ail_st_list) {
+ bh = bd->bd_bh;
+ fs_err(sdp, "bd %p: blk:0x%llx bh=%p ", bd,
+ (unsigned long long)bd->bd_blkno, bh);
+ if (!bh) {
+ fs_err(sdp, "\n");
+ continue;
+ }
+ fs_err(sdp, "0x%llx up2:%d dirt:%d lkd:%d req:%d "
+ "map:%d new:%d ar:%d aw:%d delay:%d "
+ "io err:%d unwritten:%d dfr:%d pin:%d esc:%d\n",
+ (unsigned long long)bh->b_blocknr,
+ buffer_uptodate(bh), buffer_dirty(bh),
+ buffer_locked(bh), buffer_req(bh),
+ buffer_mapped(bh), buffer_new(bh),
+ buffer_async_read(bh), buffer_async_write(bh),
+ buffer_delay(bh), buffer_write_io_error(bh),
+ buffer_unwritten(bh),
+ buffer_defer_completion(bh),
+ buffer_pinned(bh), buffer_escaped(bh));
+ fs_err(sdp, "\n");
+ }
+ }
+}
/**
* gfs2_ail1_flush - start writeback of some ail1 entries
@@ -155,11 +190,16 @@ void gfs2_ail1_flush(struct gfs2_sbd *sdp, struct writeback_control *wbc)
struct gfs2_trans *tr;
struct blk_plug plug;
int ret = 0;
+ unsigned long flush_start = jiffies;
trace_gfs2_ail_flush(sdp, wbc, 1);
blk_start_plug(&plug);
spin_lock(&sdp->sd_ail_lock);
restart:
+ if (time_after(jiffies, flush_start + (HZ * 60))) {
+ dump_ail_list(sdp);
+ goto out;
+ }
list_for_each_entry_reverse(tr, head, tr_list) {
if (wbc->nr_to_write <= 0)
break;
@@ -170,6 +210,7 @@ void gfs2_ail1_flush(struct gfs2_sbd *sdp, struct writeback_control *wbc)
break;
}
}
+out:
spin_unlock(&sdp->sd_ail_lock);
blk_finish_plug(&plug);
if (ret) {
--
2.25.1
next prev parent reply other threads:[~2020-03-26 18:40 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-03-26 18:40 [Cluster-devel] [GFS2 PATCH 0/5] gfs2; jdata io deadlocks Bob Peterson
2020-03-26 18:40 ` [Cluster-devel] [GFS2 PATCH 1/5] gfs2: don't lock sd_log_flush_lock in try_rgrp_unlink Bob Peterson
2020-03-26 18:40 ` Bob Peterson [this message]
2020-03-26 18:40 ` [Cluster-devel] [GFS2 PATCH 3/5] gfs2: change from write to read lock for sd_log_flush_lock in journal replay Bob Peterson
2020-03-26 18:40 ` [Cluster-devel] [GFS2 PATCH 4/5] gfs2: special log flush sequence to protect jdata writes Bob Peterson
2020-03-27 21:38 ` Andreas Gruenbacher
2020-03-26 18:40 ` [Cluster-devel] [GFS2 PATCH 5/5] gfs2: implement special transaction type for jdata sync writes Bob Peterson
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200326184020.123544-3-rpeterso@redhat.com \
--to=rpeterso@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).