topotests/ bgp_l3vpn_to_bgp_vrf: improvements

[louberger]

refactor to try to avoid periodic failure, also collect more info
add report zebra memory stats

127  ce1    BGPd heap: 6860 KiB --> 36 MiB (0.5828 KiB/vpn route)    1    0
128  ce1    Zebra heap: 5992 KiB --> 36 MiB (0.6002 KiB/vpn route)   1    0
129  ce2    BGPd heap: 6856 KiB --> 36 MiB (0.5829 KiB/vpn route)    1    0
130  ce2    Zebra heap: 5992 KiB --> 41 MiB (0.7002 KiB/vpn route)   1    0
131  ce3    BGPd heap: 6856 KiB --> 31 MiB (0.4829 KiB/vpn route)    1    0
132  ce3    Zebra heap: 5272 KiB --> 35 MiB (0.5946 KiB/vpn route)   1    0
133  r1     BGPd heap: 7160 KiB --> 183 MiB (3.5168 KiB/vpn route)   1    0
134  r1     Zebra heap: 7576 KiB --> 42 MiB (0.6885 KiB/vpn route)   1    0
135  r2     BGPd heap: 7016 KiB --> 169 MiB (3.2397 KiB/vpn route)   1    0
136  r2     Zebra heap: 7712 KiB --> 7712 KiB (0.0 KiB/vpn route)    1    0
137  r3     BGPd heap: 7160 KiB --> 183 MiB (3.5168 KiB/vpn route)   1    0
138  r3     Zebra heap: 7572 KiB --> 44 MiB (0.7286 KiB/vpn route)   1    0
139  r4     BGPd heap: 7300 KiB --> 275 MiB (5.354 KiB/vpn route)    1    0
140  r4     Zebra heap: 7604 KiB --> 89 MiB (1.6279 KiB/vpn route)   1    0

[mjstapp]

is the purpose of this to be a temporary change, that you'll/ we'll use to gather info, and then remove? or is it a fix for a problem (or a permanent change to report more info)?

[LabN-CI]

💚 Basic BGPD CI results: SUCCESS, 0 tests failed

Results table

_	_
Result	SUCCESS git merge/4617 4879217
Date	06/26/2019
Start	14:00:46
Finish	14:22:56
Run-Time	22:10
Total	1813
Pass	1813
Fail	0
Valgrind-Errors	0
Valgrind-Loss	0
Details	vncregress-2019-06-26-14:00:46.txt
Log	autoscript-2019-06-26-14:01:46.log.bz2
Memory	415 418 360

For details, please contact louberger

[louberger]

this is a permanent change - just another way of doing the same thing...

…

On 6/26/2019 2:14 PM, Mark Stapp wrote: is the purpose of this to be a temporary change, that you'll/ we'll use to gather info, and then remove? or is it a fix for a problem (or a permanent change to report more info)? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#4617?email_source=notifications&email_token=ABSFNSCUSALY3FKJKDF2JY3P4OWXXA5CNFSM4H3UVWR2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODYUMCZQ#issuecomment-505987430>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABSFNSENHEGR3DU2GAADJPDP4OWXXANCNFSM4H3UVWRQ>.

[NetDEF-CI]

Continuous Integration Result: SUCCESSFUL

Congratulations, this patch passed basic tests

Tested-by: NetDEF / OpenSourceRouting.org CI System

CI System Testrun URL: https://ci1.netdef.org/browse/FRR-FRRPULLREQ-8243/

This is a comment from an automated CI system.
For questions and feedback in regards to this CI system, please feel free to email
Martin Winter - mwinter (at) opensourcerouting.org.

Warnings Generated during build:

Debian 10 amd64 build: Successful with additional warnings

Debian Package lintian failed for Debian 10 amd64 build:
(see full package build log at https://ci1.netdef.org/browse/FRR-FRRPULLREQ-8243/artifact/DEB10BUILD/ErrorLog/log_lintian.txt)

W: frr source: changelog-should-mention-nmu

---

CLANG Static Analyzer Summary

• Github Pull Request 4617, comparing to Git base SHA f3afd0a

No Changes in Static Analysis warnings compared to base

1 Static Analyzer issues remaining.

See details at
https://ci1.netdef.org/browse/FRR-FRRPULLREQ-8243/artifact/shared/static_analysis/index.html

[mjstapp]

Looking at the output from the CI run. It looks as if we're logging two copies of the sho mem output, possibly? One looks formatted, then there's another set that looks like it's been stripped of newlines, maybe. that's not too useful/readable. any way to elide that?

[louberger]

The second form (without the line feeds) has been useful in seeing what goes wrong in a run.  Given our current reasonable stability of topotest, I'll turn these off.

…

On 6/26/2019 3:36 PM, Mark Stapp wrote: ***@***.**** commented on this pull request. Looking at the output from the CI run. It looks as if we're logging two copies of the |sho mem| output, possibly? One looks formatted, then there's another set that looks like it's been stripped of newlines, maybe. that's not too useful/readable. any way to elide that? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#4617?email_source=notifications&email_token=ABSFNSBRO4MN7YZNVCTQXNTP4PAMBA5CNFSM4H3UVWR2YY3PNVWWK3TUL52HS4DFWFIHK3DMKJSXC5LFON2FEZLWNFSXPKTDN5WW2ZLOORPWSZGOB4YGHYY#pullrequestreview-254829539>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABSFNSHTUV6FCBCEGTYF5YLP4PAMBANCNFSM4H3UVWRQ>.

[qlyoung]

Nit, but the correct commit label for this is tests:

[LabN-CI]

💚 Basic BGPD CI results: SUCCESS, 0 tests failed

Results table

_	_
Result	SUCCESS git merge/4617 d850e66
Date	06/26/2019
Start	16:35:41
Finish	16:57:41
Run-Time	22:00
Total	1813
Pass	1813
Fail	0
Valgrind-Errors	0
Valgrind-Loss	0
Details	vncregress-2019-06-26-16:35:41.txt
Log	autoscript-2019-06-26-16:36:40.log.bz2
Memory	432 436 360

For details, please contact louberger

[NetDEF-CI]

Continuous Integration Result: SUCCESSFUL

Congratulations, this patch passed basic tests

Tested-by: NetDEF / OpenSourceRouting.org CI System

CI System Testrun URL: https://ci1.netdef.org/browse/FRR-FRRPULLREQ-8244/

This is a comment from an automated CI system.
For questions and feedback in regards to this CI system, please feel free to email
Martin Winter - mwinter (at) opensourcerouting.org.

Warnings Generated during build:

Debian 10 amd64 build: Successful with additional warnings

Debian Package lintian failed for Debian 10 amd64 build:
(see full package build log at https://ci1.netdef.org/browse/FRR-FRRPULLREQ-8244/artifact/DEB10BUILD/ErrorLog/log_lintian.txt)

W: frr source: changelog-should-mention-nmu

---

CLANG Static Analyzer Summary

• Github Pull Request 4617, comparing to Git base SHA f3afd0a

No Changes in Static Analysis warnings compared to base

1 Static Analyzer issues remaining.

See details at
https://ci1.netdef.org/browse/FRR-FRRPULLREQ-8244/artifact/shared/static_analysis/index.html

[mjstapp]

One of the issues we've had with this suite has been the timeout in the scale_down test. it had seemed as if test was computing a timeout to use, but then failing after a shorter delay. it looks as if that hasn't been changed - have you come to some conclusion about what was happening there?

[louberger]

no as could not reproduce -- after 200+ runs!.  I'm speculating that there was some condition that resulted in grep -c returning '0' - so changed command to not use grep -c.

…

On 6/27/2019 9:26 AM, Mark Stapp wrote: ***@***.**** commented on this pull request. One of the issues we've had with this suite has been the timeout in the scale_down test. it had seemed as if test was computing a timeout to use, but then failing after a shorter delay. it looks as if that hasn't been changed - have you come to some conclusion about what was happening there? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#4617?email_source=notifications&email_token=ABSFNSGZORMZQYG7PL6KP73P4S5XFA5CNFSM4H3UVWR2YY3PNVWWK3TUL52HS4DFWFIHK3DMKJSXC5LFON2FEZLWNFSXPKTDN5WW2ZLOORPWSZGOB43GQYQ#pullrequestreview-255223906>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABSFNSDD6MGCPNYVZB5ML5LP4S5XFANCNFSM4H3UVWRQ>.

[mjstapp]

Looks good - hopefully this'll make the suite less fragile.