mirror of
https://codeberg.org/redict/redict.git
synced 2025-01-23 08:38:27 -05:00
45a155bd0f
To avoid data loss, this commit adds a grace period for lagging replicas to
catch up the replication offset.
Done:
* Wait for replicas when shutdown is triggered by SIGTERM and SIGINT.
* Wait for replicas when shutdown is triggered by the SHUTDOWN command. A new
blocked client type BLOCKED_SHUTDOWN is introduced, allowing multiple clients
to call SHUTDOWN in parallel.
Note that they don't expect a response unless an error happens and shutdown is aborted.
* Log warning for each replica lagging behind when finishing shutdown.
* CLIENT_PAUSE_WRITE while waiting for replicas.
* Configurable grace period 'shutdown-timeout' in seconds (default 10).
* New flags for the SHUTDOWN command:
- NOW disables the grace period for lagging replicas.
- FORCE ignores errors writing the RDB or AOF files which would normally
prevent a shutdown.
- ABORT cancels ongoing shutdown. Can't be combined with other flags.
* New field in the output of the INFO command: 'shutdown_in_milliseconds'. The
value is the remaining maximum time to wait for lagging replicas before
finishing the shutdown. This field is present in the Server section **only**
during shutdown.
Not directly related:
* When shutting down, if there is an AOF saving child, it is killed **even** if AOF
is disabled. This can happen if BGREWRITEAOF is used when AOF is off.
* Client pause now has end time and type (WRITE or ALL) per purpose. The
different pause purposes are *CLIENT PAUSE command*, *failover* and
*shutdown*. If clients are unpaused for one purpose, it doesn't affect client
pause for other purposes. For example, the CLIENT UNPAUSE command doesn't
affect client pause initiated by the failover or shutdown procedures. A completed
failover or a failed shutdown doesn't unpause clients paused by the CLIENT
PAUSE command.
Notes:
* DEBUG RESTART doesn't wait for replicas.
* We already have a warning logged when a replica disconnects. This means that
if any replica connection is lost during the shutdown, it is either logged as
disconnected or as lagging at the time of exit.
Co-authored-by: Oran Agra <oran@redislabs.com>
69 lines
2.2 KiB
Tcl
69 lines
2.2 KiB
Tcl
start_server {tags {"shutdown external:skip"}} {
|
|
test {Temp rdb will be deleted if we use bg_unlink when shutdown} {
|
|
for {set i 0} {$i < 20} {incr i} {
|
|
r set $i $i
|
|
}
|
|
# It will cost 2s(20 * 100ms) to dump rdb
|
|
r config set rdb-key-save-delay 100000
|
|
|
|
# Child is dumping rdb
|
|
r bgsave
|
|
after 100
|
|
set dir [lindex [r config get dir] 1]
|
|
set child_pid [get_child_pid 0]
|
|
set temp_rdb [file join [lindex [r config get dir] 1] temp-${child_pid}.rdb]
|
|
# Temp rdb must be existed
|
|
assert {[file exists $temp_rdb]}
|
|
|
|
catch {r shutdown nosave}
|
|
# Make sure the server was killed
|
|
catch {set rd [redis_deferring_client]} e
|
|
assert_match {*connection refused*} $e
|
|
|
|
# Temp rdb file must be deleted
|
|
assert {![file exists $temp_rdb]}
|
|
}
|
|
}
|
|
|
|
start_server {tags {"shutdown external:skip"}} {
|
|
test {SHUTDOWN ABORT can cancel SIGTERM} {
|
|
r debug pause-cron 1
|
|
set pid [s process_id]
|
|
exec kill -SIGTERM $pid
|
|
after 10; # Give signal handler some time to run
|
|
r shutdown abort
|
|
verify_log_message 0 "*Shutdown manually aborted*" 0
|
|
r debug pause-cron 0
|
|
r ping
|
|
} {PONG}
|
|
|
|
test {Temp rdb will be deleted in signal handle} {
|
|
for {set i 0} {$i < 20} {incr i} {
|
|
r set $i $i
|
|
}
|
|
# It will cost 2s (20 * 100ms) to dump rdb
|
|
r config set rdb-key-save-delay 100000
|
|
|
|
set pid [s process_id]
|
|
set temp_rdb [file join [lindex [r config get dir] 1] temp-${pid}.rdb]
|
|
|
|
# trigger a shutdown which will save an rdb
|
|
exec kill -SIGINT $pid
|
|
# Wait for creation of temp rdb
|
|
wait_for_condition 50 10 {
|
|
[file exists $temp_rdb]
|
|
} else {
|
|
fail "Can't trigger rdb save on shutdown"
|
|
}
|
|
|
|
# Insist on immediate shutdown, temp rdb file must be deleted
|
|
exec kill -SIGINT $pid
|
|
# wait for the rdb file to be deleted
|
|
wait_for_condition 50 10 {
|
|
![file exists $temp_rdb]
|
|
} else {
|
|
fail "Can't trigger rdb save on shutdown"
|
|
}
|
|
}
|
|
}
|