[master] 1b0a51cc1 test: Over-breeding is still a thing

Dridi Boukelmoune dridi.boukelmoune at gmail.com
Wed Oct 28 09:48:08 UTC 2020


commit 1b0a51cc1ff3890ea1ff1949fec7fd9297f3735e
Author: Dridi Boukelmoune <dridi.boukelmoune at gmail.com>
Date:   Wed Oct 28 10:44:37 2020 +0100

    test: Over-breeding is still a thing
    
    Spotted on CircleCI:
    
        ---- v1    Not true: MAIN.threads (26) == 10 (10)
    
    Refs #2942

diff --git a/bin/varnishtest/tests/v00006.vtc b/bin/varnishtest/tests/v00006.vtc
index 66d3d989c..89bf9a619 100644
--- a/bin/varnishtest/tests/v00006.vtc
+++ b/bin/varnishtest/tests/v00006.vtc
@@ -23,10 +23,10 @@ varnish v1 \
 # >thread_pool_timeout to allow for any over-bred workers to be kissed
 # to death.
 # NB: This is important for to avoid mis-ordering of the workers.
-
 delay 11
 
-varnish v1 -expect MAIN.threads == 10
+# But all in all, the delay does not fully prevent over-breeding.
+varnish v1 -expect MAIN.threads >= 10
 
 client c1 {
 	txreq -url "/bar"


More information about the varnish-commit mailing list