ecda894264
* add patch
boo1235399-fix_the_warning_of_blk_validate_limits.patch
OBS-URL: https://build.opensuse.org/package/show/network:ha-clustering:Factory/drbd?expand=0&rev=156
120 lines
4.8 KiB
Diff
120 lines
4.8 KiB
Diff
From aab03bfc73a62f95011316545a5c0fbb4817741b Mon Sep 17 00:00:00 2001
|
|
From: Lars Ellenberg <lars.ellenberg@linbit.com>
|
|
Date: Wed, 14 Aug 2024 11:49:42 +0200
|
|
Subject: [PATCH 01/32] drbd: properly rate-limit resync progress reports
|
|
|
|
A peer_device in "paused" sync would have flooded the "drbd events2"
|
|
generic netlink broadcast with "resync progress reports",
|
|
if it cleared significant out-of-sync bits,
|
|
as is the case with application writes,
|
|
or several peers syncing from the same sync source
|
|
and having a "paused sync" replication state between themselves.
|
|
|
|
If you have "many" such resources, this storm may even overflow receive buffers.
|
|
At most one progress report every three seconds should be enough,
|
|
and is what was intended.
|
|
|
|
Use a new "last progress report time stamp" to throttle
|
|
advancing resync progress marks and progress report broadcasts.
|
|
---
|
|
drbd/drbd_actlog.c | 35 +++++++++++++++++++++++------------
|
|
drbd/drbd_int.h | 1 +
|
|
drbd/drbd_receiver.c | 1 +
|
|
drbd/drbd_state.c | 2 ++
|
|
4 files changed, 27 insertions(+), 12 deletions(-)
|
|
|
|
diff --git a/drbd/drbd_actlog.c b/drbd/drbd_actlog.c
|
|
index b96560843878..646dcb29e1d9 100644
|
|
--- a/drbd/drbd_actlog.c
|
|
+++ b/drbd/drbd_actlog.c
|
|
@@ -1020,19 +1020,30 @@ static bool update_rs_extent(struct drbd_peer_device *peer_device,
|
|
|
|
void drbd_advance_rs_marks(struct drbd_peer_device *peer_device, unsigned long still_to_go)
|
|
{
|
|
- unsigned long now = jiffies;
|
|
- unsigned long last = peer_device->rs_mark_time[peer_device->rs_last_mark];
|
|
- int next = (peer_device->rs_last_mark + 1) % DRBD_SYNC_MARKS;
|
|
- if (time_after_eq(now, last + DRBD_SYNC_MARK_STEP)) {
|
|
- if (peer_device->rs_mark_left[peer_device->rs_last_mark] != still_to_go &&
|
|
- peer_device->repl_state[NOW] != L_PAUSED_SYNC_T &&
|
|
- peer_device->repl_state[NOW] != L_PAUSED_SYNC_S) {
|
|
- peer_device->rs_mark_time[next] = now;
|
|
- peer_device->rs_mark_left[next] = still_to_go;
|
|
- peer_device->rs_last_mark = next;
|
|
- }
|
|
- drbd_peer_device_post_work(peer_device, RS_PROGRESS);
|
|
+ unsigned long now;
|
|
+ int next;
|
|
+
|
|
+ /* report progress and advance marks only if we made progress */
|
|
+ if (peer_device->rs_mark_left[peer_device->rs_last_mark] == still_to_go)
|
|
+ return;
|
|
+
|
|
+ /* report progress and advance marks at most once every DRBD_SYNC_MARK_STEP (3 seconds) */
|
|
+ now = jiffies;
|
|
+ if (!time_after_eq(now, peer_device->rs_last_progress_report_ts + DRBD_SYNC_MARK_STEP))
|
|
+ return;
|
|
+
|
|
+ /* Do not advance marks if we are "paused" */
|
|
+ if (peer_device->repl_state[NOW] != L_PAUSED_SYNC_T &&
|
|
+ peer_device->repl_state[NOW] != L_PAUSED_SYNC_S) {
|
|
+ next = (peer_device->rs_last_mark + 1) % DRBD_SYNC_MARKS;
|
|
+ peer_device->rs_mark_time[next] = now;
|
|
+ peer_device->rs_mark_left[next] = still_to_go;
|
|
+ peer_device->rs_last_mark = next;
|
|
}
|
|
+
|
|
+ /* But still report progress even if paused. */
|
|
+ peer_device->rs_last_progress_report_ts = now;
|
|
+ drbd_peer_device_post_work(peer_device, RS_PROGRESS);
|
|
}
|
|
|
|
/* It is called lazy update, so don't do write-out too often. */
|
|
diff --git a/drbd/drbd_int.h b/drbd/drbd_int.h
|
|
index 49bd7b0c407c..c18407899f59 100644
|
|
--- a/drbd/drbd_int.h
|
|
+++ b/drbd/drbd_int.h
|
|
@@ -1285,6 +1285,7 @@ struct drbd_peer_device {
|
|
unsigned long rs_paused;
|
|
/* skipped because csum was equal [unit BM_BLOCK_SIZE] */
|
|
unsigned long rs_same_csum;
|
|
+ unsigned long rs_last_progress_report_ts;
|
|
#define DRBD_SYNC_MARKS 8
|
|
#define DRBD_SYNC_MARK_STEP (3*HZ)
|
|
/* block not up-to-date at mark [unit BM_BLOCK_SIZE] */
|
|
diff --git a/drbd/drbd_receiver.c b/drbd/drbd_receiver.c
|
|
index 19634f6423bd..ee54cf3ac116 100644
|
|
--- a/drbd/drbd_receiver.c
|
|
+++ b/drbd/drbd_receiver.c
|
|
@@ -3409,6 +3409,7 @@ static int receive_DataRequest(struct drbd_connection *connection, struct packet
|
|
peer_device->ov_skipped = 0;
|
|
peer_device->rs_total = ov_left;
|
|
peer_device->rs_last_writeout = now;
|
|
+ peer_device->rs_last_progress_report_ts = now;
|
|
for (i = 0; i < DRBD_SYNC_MARKS; i++) {
|
|
peer_device->rs_mark_left[i] = ov_left;
|
|
peer_device->rs_mark_time[i] = now;
|
|
diff --git a/drbd/drbd_state.c b/drbd/drbd_state.c
|
|
index be1de8f0653b..44f55ee5c939 100644
|
|
--- a/drbd/drbd_state.c
|
|
+++ b/drbd/drbd_state.c
|
|
@@ -2483,6 +2483,7 @@ static void initialize_resync_progress_marks(struct drbd_peer_device *peer_devic
|
|
unsigned long now = jiffies;
|
|
int i;
|
|
|
|
+ peer_device->rs_last_progress_report_ts = now;
|
|
for (i = 0; i < DRBD_SYNC_MARKS; i++) {
|
|
peer_device->rs_mark_left[i] = tw;
|
|
peer_device->rs_mark_time[i] = now;
|
|
@@ -2730,6 +2731,7 @@ static void finish_state_change(struct drbd_resource *resource, const char *tag)
|
|
peer_device->ov_last_skipped_size = 0;
|
|
peer_device->ov_last_skipped_start = 0;
|
|
peer_device->rs_last_writeout = now;
|
|
+ peer_device->rs_last_progress_report_ts = now;
|
|
for (i = 0; i < DRBD_SYNC_MARKS; i++) {
|
|
peer_device->rs_mark_left[i] = peer_device->rs_total;
|
|
peer_device->rs_mark_time[i] = now;
|
|
--
|
|
2.35.3
|
|
|