upstream/haproxy
1
0
Fork 0
mirror of https://github.com/haproxy/haproxy.git synced 2026-02-03 20:39:41 -05:00
Code Issues Projects Releases Packages Wiki Activity Actions 10

DOC: internals: cleanup few typos in master-worker documentation
Some checks are pending
Contrib / build (push) Waiting to run
Details
alpine/musl / gcc (push) Waiting to run
Details
VTest / Generate Build Matrix (push) Waiting to run
Details
VTest / (push) Blocked by required conditions
Details
Windows / Windows, gcc, all features (push) Waiting to run
Details

Browse source 
s/mecanism/mechanism
s/got ride/got rid
s/traditionnal/traditional

One typo is confusion between master and worker that results to a
semantic mistake in the sentence:

"...the master will emit an "exit-on-failure" error and will kill every
workers with a SIGTERM and exits with the same error code than the
failed [-master-]{+worker+}..."

Should be backported as far as 3.1.
This commit is contained in:
Egor Shestakov 2026-01-29 17:15:33 +00:00 committed by William Lallemand
parent da728aa0f6
commit 02e6375017
1 changed files with 5 additions and 5 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

10
doc/internals/mworker.md
View file

@ -11,7 +11,7 @@ default init, this was controversial but fedora and archlinux already uses it.
At this time HAProxy still had a multi-process model, and the way haproxy is
working was incompatible with the daemon mode.
Systemd is compatible with traditionnal forking services, but somehow HAProxy
Systemd is compatible with traditional forking services, but somehow HAProxy
is different. To work correctly, systemd needs a main PID, this is the PID of
the process that systemd will supervises.
@ -45,7 +45,7 @@ However the wrapper suffered from several problems:
### mworker V1
HAProxy 1.8 got ride of the wrapper which was replaced by the master worker
HAProxy 1.8 got rid of the wrapper which was replaced by the master worker
mode. This first version was basically a reintegration of the wrapper features
within HAProxy. HAProxy is launched with the -W flag, read the configuration and
then fork. In mworker mode, the master is usually launched as a root process,
@ -86,7 +86,7 @@ retrieved automatically.
The master is supervising the workers, when a current worker (not a previous one
from before the reload) is exiting without being asked for a reload, the master
will emit an "exit-on-failure" error and will kill every workers with a SIGTERM
and exits with the same error code than the failed master, this behavior can be
and exits with the same error code than the failed worker, this behavior can be
changed by using the "no exit-on-failure" option in the global section.
While the master is supervising the workers using the wait() function, the
@ -186,8 +186,8 @@ number that can be found in HAPROXY_PROCESSES. With this change the stats socket
in the configuration is less useful and everything can be done from the master
CLI.
With 2.7, the reload mecanism of the master CLI evolved, with previous versions,
this mecanism was asynchronous, so once the `reload` command was received, the
With 2.7, the reload mechanism of the master CLI evolved, with previous versions,
this mechanism was asynchronous, so once the `reload` command was received, the
master would reload, the active master CLI connection was closed, and there was
no way to return a status as a response to the `reload` command. To achieve a
synchronous reload, a dedicated sockpair is used, one side uses a master CLI