Skip to contentSkip to conversation

osstest: increase boot timeout for Debian PV guests

PATCH In review Roger Pau Monne 3 participants 5 messages Apr 12, 2024 – Apr 15, 2024

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 osstest: increase boot timeout for Debian PV guests

The current timeout of 40s seems to be too low for AMD boxes (pinots and rimavas) in the lab after XSA-455, see:

http://logs.test-lab.xenproject.org/osstest/logs/185303/test-amd64-coresched-amd64-xl/info.html

Increase the timeout to 60s.

---
 ts-debian-di-install | 2 +-
 ts-debian-install    | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
ts-debian-di-install +1 -1
index 06c7e1f46a92..1a187dfeb9cb 100755
@@ -76,7 +76,7 @@ sub prep () {
     target_install_packages_norec($ho, qw(lvm2));
 
     $gho= prepareguest($ho, $gn, $guesthost, 22,
-                       $disk_mb, 40);
+                       $disk_mb, 60);
 
     prepareguest_part_diskimg($ho, $gho, $disk_mb);
 }
ts-debian-install +1 -1
index 62db487ad15d..ef2954dd3c8e 100755
@@ -42,7 +42,7 @@ sub prep () {
 
     $gho= prepareguest($ho, $gn, $guesthost, 22,
                        $swap_mb + $disk_mb + 2,
-                       40);
+                       60);
     target_cmd_root($ho, "umount $gho->{Lvdev} ||:");
 }
 
Signature · 1 line
2.44.0
#2 lore raw
In reply to Roger Pau Monne
Quoted 2 lines · Roger Pau Monne

The current timeout of 40s seems to be too low for AMD boxes (pinots and rimavas) in the lab after XSA-455, see:

There's something else we can tweak if only some machine need extra time, it is an host property "TimeoutFactor", which can increase all timeout for a single machine. (It's use on the cubietruck for example.)

Or is it better to just increase boot time for all (or at least those) pv guest?

Cheers,

Signature · 1 line
Anthony PERARD
#3 lore raw
Quoted 10 lines · Anthony PERARD
Quoted 2 lines · Roger Pau Monne

The current timeout of 40s seems to be too low for AMD boxes (pinots and rimavas) in the lab after XSA-455, see:

There's something else we can tweak if only some machine need extra time, it is an host property "TimeoutFactor", which can increase all timeout for a single machine. (It's use on the cubietruck for example.)

Or is it better to just increase boot time for all (or at least those) pv guest?

I did consider that, but given the timeout is just limited to PV guest startup I considered the "TimeoutFactor" too broad. I think increasing the Debian PV boot timeout from 40s to 60s is a minor adjustment, and shouldn't affect other tests.

Let me know if you still prefer to use "TimeoutFactor" and I will look into it.

Thanks, Roger.

#4 lore raw
Quoted 12 lines · Roger Pau Monné
Quoted 10 lines · Anthony PERARD
Quoted 2 lines · Roger Pau Monne

The current timeout of 40s seems to be too low for AMD boxes (pinots and rimavas) in the lab after XSA-455, see:

There's something else we can tweak if only some machine need extra time, it is an host property "TimeoutFactor", which can increase all timeout for a single machine. (It's use on the cubietruck for example.)

Or is it better to just increase boot time for all (or at least those) pv guest?

Quoted 7 lines

I did consider that, but given the timeout is just limited to PV guest startup I considered the "TimeoutFactor" too broad. I think increasing the Debian PV boot timeout from 40s to 60s is a minor adjustment, and shouldn't affect other tests.

Let me know if you still prefer to use "TimeoutFactor" and I will look into it.

Thanks,

Signature · 1 line
Anthony PERARD