From 01183a77aa9fabe4f949bad60d4f27f067737e1b Mon Sep 17 00:00:00 2001 From: Tomas Volf <~@wolfsden.cz> Date: Fri, 5 Jul 2024 17:01:20 +0200 Subject: [PATCH] services: web: Pass run-directory to nginx's actions. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Reloading nginx lead to (harmless, yet annoying) alert being logged: nginx: [alert] could not open error log file: open() "/gnu/store/9l2lznlqpjcm79znq1xas378maqgysc8-nginx-1.27.0/logs/error.log" failed (2: No such file or directory) Since there already is prepared runtime directory for nginx, all that was missing was to also pass it to the reload and reopen invocations. * gnu/services/web.scm (nginx-shepherd-service)[reload, reopen]: Pass -p flag with run-directory as value. Change-Id: Id5c558d65def8c1fe5cf581a4a370508ac05e550 Signed-off-by: Ludovic Courtès --- gnu/services/web.scm | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/gnu/services/web.scm b/gnu/services/web.scm index ee3499e5cd..c434436f69 100644 --- a/gnu/services/web.scm +++ b/gnu/services/web.scm @@ -857,11 +857,11 @@ This has the effect of killing old worker processes and starting new ones, using the same configuration file. It is useful for situations where the same nginx configuration file can point to different things after a reload, such as renewed TLS certificates, or @code{include}d files.") - (procedure (nginx-action "-s" "reload"))) + (procedure (nginx-action "-p" run-directory "-s" "reload"))) (shepherd-action (name 'reopen) (documentation "Re-open log files.") - (procedure (nginx-action "-s" "reopen")))))))))) + (procedure (nginx-action "-p" run-directory "-s" "reopen")))))))))) (define nginx-service-type (service-type (name 'nginx)