[Swan-dev] new test failures

[D. Hugh Redelmeier]

| From: Paul Wouters <paul at nohats.ca>

Thanks for looking at these.

| On Wed, 13 Feb 2019, D. Hugh Redelmeier wrote:
| 
| > I ran the tests last evening.  The new failures look simple to fix.
| >
| > I include a diff of the summary from the previous run
| > (summary produced by "testing/utils/kvmresults.py testing/pluto/").
| 
| I'm going to ignore the changes due to andrew's algorithm changes from
| the last few days, assuming those will be updated.

Right.  We'll leave it to Andrew.

| > - testing/pluto/ikev2-liveness-09 failed in its mysterious way:
| > -| xfrm acquire rtattribute type 5
| > -| xfrm acquire rtattribute type 16
| 
|  grep -E "liveness: action|acquire" OUTPUT/west.pluto.log
| 
| that's an odd grep line. I think the case is really only looking
| for:
| 
| 	initiate on demand from 192.0.1.254:8 to 192.0.2.254:0 proto=1
| because: acquire

Should the grep be revised?

| > - testing/pluto/nat-pluto-04 failed in another mysterious way
| > +? (192.139.70.77) at fe:00:00:dc:bc:ff [ether] on eth0
| 
| NetName:        MIMOSAL
| 
| That IP is your LAN. Why is it leaking into the tests? :)

Yeah: why?

I don't know of anything I've done to cause this.

| >> testing/pluto/ikev2-child-03-dh-none-pfs-no failed west:output-different
| > testing/pluto/ikev2-child-03-dh-none-pfs-no/OUTPUT/west.console.diff
| > -004 "westnet-eastnet-ikev2" #2: STATE_V2_IPSEC_I: IPsec SA established
| > tunnel mode {ESP=>0xESPESP <0xESPESP xfrm=AES_CBC_128-HMAC_SHA1_96
| > NATOA=none NATD=none DPD=passive}
| > +004 "westnet-eastnet-ikev2" #2: STATE_V2_IPSEC_I: IPsec SA established
| > tunnel mode {ESP=>0xESPESP <0xESPESP xfrm=AES_CBC_256-HMAC_SHA2_512_256
| > NATOA=none NATD=none DPD=passive}
| 
| Ah few cases got their default key size to go from 128 to 256? Probably
| as a result of the proposal parser changes? I'm fine with that. Let uses
| go back to 128 key manually if they really want to do that.

Why intervene?  If the code prefers 256, isn't that what's best
tested?  Of course a few 128 tests are in order.

Again, it looks as if the reference logs need to be updated.  Probably
by Andrew.