All new accounts created on Gitlab now require administrator approval. If you invite any collaborators, please let Flux staff know so they can approve the accounts.

Commit 6bdba92c authored by Mac Newbold's avatar Mac Newbold

Decrease the sleep between loops from 2 to 1, and fix a typo. This should

help nodes in reload_pending get sucked into reloading faster. If it
doesn't do enough, we'll need to do more batching of stuff, so we get some
parallelism in os_load instead of forcing it to serialize by calling
os_load one node at a time.

I was tempted to nuke all the stuff that was in there from the netdisk
reload type, but decided not to. It won't be too long (relatively
speaking) before we have freed, the new "free node manager" that will
replace/supersede our current reload_daemon anyway.
parent a77a1559
......@@ -103,7 +103,7 @@ while (1) {
#
# Partial delay between loops in case of an error.
#
sleep(2);
sleep(1);
#
# We use this to figure out when to delete nodes from the retried and
......@@ -261,7 +261,7 @@ while (1) {
#
# This should not fail!
#
fatal("$os_load failed on $node. Thats not supposed to happen.\n".
fatal("$os_load failed on $node. That's not supposed to happen.\n".
"Please check the reload daemon log before restarting!\n");
}
}
......@@ -298,6 +298,10 @@ while (1) {
next;
#}
#
# START OBSOLETE SECTION
#
#
# Reload was started. We want to wait until its finished.
#
......@@ -347,6 +351,9 @@ while (1) {
goto again;
}
sleep(10);
#
# END OBSOLETE SECTION
#
}
sub fatal($)
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment