-
Notifications
You must be signed in to change notification settings - Fork 21
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Release 2020.04 - RC1 #152
Comments
[10-icmpv6-error] Task 01: PASS
[10-icmpv6-error] Task 02: PASS
[10-icmpv6-error] Task 03: PASS
[10-icmpv6-error] Task 04: PASS
[10-icmpv6-error] Task 05: PASS
[10-icmpv6-error] Task 06: PASS
|
[10-icmpv6-error] Task 08: PASS
[10-icmpv6-error] Task 09: PASS
[10-icmpv6-error] Task 10: PASS
|
11-lorawan Task 1 Passes
11-lorawan Task 2 Passes
11-lorawan Task 3 Passes
11-lorawan Task 4 Passes
|
04-single-hop-6lowpan-icmp Task 1 Passes
04-single-hop-6lowpan-icmp Task 2 Passes
04-single-hop-6lowpan-icmp Task 3 Passes
04-single-hop-6lowpan-icmp Task 4 Passes
04-single-hop-6lowpan-icmp Task 9 Passes
04-single-hop-6lowpan-icmp Task 10 Passes
|
[08-interop] Task 11: PASS
|
06-single-hop-udp Task 3 Passes
06-single-hop-udp Task 4 Passes
06-single-hop-udp Task 5 Passes
06-single-hop-udp Task 6 Passes
|
[04-single-hop-6lowpan-icmp] Task 05: PASS
[04-single-hop-6lowpan-icmp] Task 06: PASS
|
03-single-hop-ipv6-icmp Task 1 Passes
03-single-hop-ipv6-icmp Task 2 Passes
03-single-hop-ipv6-icmp Task 3 Passes
03-single-hop-ipv6-icmp Task 6 Passes
|
[09-coap] Task 01: PASS
[09-coap] Task 03: PASS
[09-coap] Task 04: PASS
[09-coap] Task 05: PASS
|
[08-interop] Task 01: PASS
|
05-single-hop-route Task 1 Passes
05-single-hop-route Task 2 Passes
05-single-hop-route Task 3 Passes
05-single-hop-route Task 4 Passes
|
[08-interop] Task 08: PASS
[08-interop] Task 09: PASS
[08-interop] Task 10: PASS
|
[02-tests] Task 01: PASS |
06-single-hop-udp Task 2 Passes
|
06-single-hop-udp Task 1 Passes
|
[10-icmpv6-error] Task 7: PASS
|
03-single-hop-ipv6-icmp Task 4 Passes
|
[08-interop] Task 02: PASS
|
03-single-hop-ipv6-icmp Task 5 FAILS This seems to be expected because of RIOT-OS/RIOT#12565, but differently than what was pointed out there the leak is occurring in the pinged node, and not the pinger. I tried multiple times with the same result, pingers pktbuf is empty and pinged node pktbuf tends to leak (not always). I also did a run with 10x the amount of pings, with the same results. Some of the pinger nodes hanged for a long time (tens of minutes) but at some point returned to normal with empty packet buffers. @miri64 have there been any changes that could explain this shift of the leak to the pinged node? tap0 OK
tap1 OK
tap2 OK
tap3 OK
tap4 OK
tap5 OK
tap6 OK
tap7 OK
tap8 OK
tap9 OK
tap10 (pinged node) LEAK
|
If there is a leak on the pinged node this is a bug. The only change that comes to mind which was also making problems in other tests already is RIOT-OS/RIOT#12994. Can you confirm that it works before that one was merged? |
Other than that: You could use RIOT-OS/RIOT#13221 to determine the make-up of the packets remaining in the packet buffer (a good excuse to test that ;-)). This could help greatly to narrow down the area of code the leak might occur. |
It is of course also a bug if the pinging node has leaks, but a leak on a pinged node is more drastic, as it can be used as an attack surface for a DoS attack. |
Well I've tried to reproduce to be able to compare, and now I'm unable to reproduce my earlier results... I'll keep trying, and get back to you. |
I think I might not have given the |
The pinged buffer had remaining packets for about 10mins after pinging was over, but it emptied after that time. I saw a leak on one of the pingers this time, and it hasn't emptied so far. |
on the pinge_rs_ or the pinge_d_ node? If its the pingers, its |
@fjmolinas please confirm: this is exactly as reported in RIOT-OS/RIOT#12565? If yes, I won't investigate into this, as I simply don't have the time right now :-(. It is normal that some packets stay around for some while and then disappear from the packet buffer, as they might be stuck in buffers, queues, etc. while waiting for other events or a timeout. |
I now have the impression that yes, it seems the first times I saw it I didn't wait enough for the packet buffer to empty, since the last couple of times I saw a packet linger, it emptied after 5-10 mins. I'll let you know If I have reasons to think its not as RIOT-OS/RIOT#12565. I'll update the specs as well on these tests, since it seems the last couple of releases we have though there was an issue because of these later-lingering packets (but that do eventually disappear). |
[07-multi-hop] Task 03: PASS
[07-multi-hop] Task 04: PASS
|
[07-multi-hop] Task 05: FAIL (see #142 (comment)) |
[01-ci] Task 02: PASS
[01-ci] Task 03: PASS
[01-ci] Task 04: PASS
|
[08-interopt] Task 04: PASSRiot Node:
Linux node (Internet Host):
|
[08-interop] Task 03: PASS
|
I don't have the output anymore, I'll try to do it again later today and parse it. |
Output of
I think most of them are unrelated stuff. Anyway, I will dig deeper |
[02-tests] Task 03
Edit: A fix for this has been provided in RIOT-OS/RIOT#13882, and backported the release branch in RIOT-OS/RIOT#13883. This task should be rerun on RC2. |
[08-interop] Task 05: PASS
|
[08-interop] Task 06: PASSNode:
Linux host:
[08-interop] Task 07: PASSNode:
Linux Host:
|
[99-compile-and-test-one-board]
cc2650-launchpad/failuresummary.mdFailures during compilation: Failures during test:
ek-lm4f120xl/failuresummary.mdFailures during compilation: Failures during test:
frdm-k64f/failuresummary.mdFailures during compilation: Failures during test:
msba2/failuresummary.mdFailures during compilation: Failures during test:
|
These are the results of the compile test: RESULTS
All manually run buildtest again there and there seems to be just "insufficient memory". The |
@RIOT-OS/maintainers is there someone who has access to the hardware to run Spec 04- Task 07 ICMPv6 multicast echo with samr21-xpro/zero + xbee and Task 08 - ICMPv6 echo with samr21-xpro/zero + xbee? A samr21-xpro and an arduino-zero with xbee shield. |
Both are available in the IoT-LAB. |
For some reason it's not working for me to book the arduino nodes |
I don't have a |
Thanks a lot! |
04-single-hop-6lowpan-icmp Task 7 Passes on iotlab-m3 and arduino-zero+xbee
04-single-hop-6lowpan-icmp Task 8 Passes on iotlab-m3 and arduino-zero+xbee
Depending on the xbee shield RIOT-OS/RIOT#13224 is needed. |
Note only needed because of the |
I'll be creating RC2 when RIOT-OS/RIOT#13922 is in. |
Closing in favor of #153 |
Thanks all for helping out! |
This issue lists the status of all tests for the Release Candidate 1 of the 2020.04 release.
Specs tested:
The text was updated successfully, but these errors were encountered: