Uh oh!
There was an error while loading. Please reload this page.
vpc: fix ips on wrong interfaces after rebooting vpc vrs - #4467
Conversation
| nic.setDeviceId(deviceId); | ||
| _nicDao.update(nic.getId(), nic); | ||
| return nic; | ||
| } |
There was a problem hiding this comment.
method name is not clear, Either define & use separate methods (for "updating nic device id" and "getting nic with device id"), or change the name appropriately.
There was a problem hiding this comment.
getNicAndUpdtaeDeviceId() might have been more accurate but this is clearly stating what it does on the tin (to me) , Can you be more explicit about what you want @sureshanaparti ?
There was a problem hiding this comment.
getNicAndUpdtaeDeviceId()might have been more accurate but this is clearly stating what it does on the tin (to me) , Can you be more explicit about what you want @sureshanaparti ?
simply "updateNicWithDeviceId()" returning nic would be ok.
There was a problem hiding this comment.
@sureshanapartiupdateNicWithDeviceId() looks good to me. @DaanHoogland what's your opinion ?
There was a problem hiding this comment.
yes, fine. the only problem i have is with my own spelling; getNicAndUpdtaeDeviceId() misses a 'k' ;)
There was a problem hiding this comment.
@sureshanaparti@DaanHoogland thanks guys. code is changed.
DaanHoogland
commented
Nov 13, 2020
@blueorangutan package |
blueorangutan
commented
Nov 13, 2020
@DaanHoogland a Jenkins job has been kicked to build packages. I'll keep you posted as I make progress. |
blueorangutan
commented
Nov 13, 2020
Packaging result: ✔centos7 ✖centos8 ✔debian. JID-2382 |
DaanHoogland
commented
Nov 13, 2020
@blueorangutan test |
blueorangutan
commented
Nov 13, 2020
@DaanHoogland a Trillian-Jenkins test job (centos7 mgmt + kvm-centos7) has been kicked to run smoke tests |
blueorangutan
commented
Nov 13, 2020
Trillian test result (tid-3169)
|
weizhouapache
commented
Nov 15, 2020
@DaanHoogland please do not merge this pr. I found an issue which might be caused by this pr. |
DaanHoogland
commented
Nov 17, 2020
@weizhouapache converted to draft so it is clear 4.15 is not waiting for it. please change back if you disagree |
…unning or configuration has changed
weizhouapache
commented
Nov 17, 2020
@DaanHoogland fixed a bug by commit f603816 in description of commit 7988f51, it says that "Force a restart of keepalived if conntrackd is not running or configuration has changed". however, in code, it is "and" not "or" |
DaanHoogland
commented
Nov 18, 2020
@weizhouapache can you add soem test data/results? |
yadvr
commented
Nov 18, 2020
@blueorangutan package |
blueorangutan
commented
Nov 18, 2020
@rhtyd a Jenkins job has been kicked to build packages. I'll keep you posted as I make progress. |
sureshanaparti
left a comment
There was a problem hiding this comment.
changes LGTM, haven't tested.
blueorangutan
commented
Nov 18, 2020
Packaging result: ✔centos7 ✖centos8 ✔debian. JID-2396 |
DaanHoogland
commented
Nov 18, 2020
@blueorangutan package |
blueorangutan
commented
Nov 18, 2020
@DaanHoogland a Jenkins job has been kicked to build packages. I'll keep you posted as I make progress. |
blueorangutan
commented
Nov 18, 2020
Packaging result: ✔centos7 ✖centos8 ✔debian. JID-2399 |
DaanHoogland
commented
Nov 18, 2020
@blueorangutan test |
blueorangutan
commented
Nov 18, 2020
@DaanHoogland a Trillian-Jenkins test job (centos7 mgmt + kvm-centos7) has been kicked to run smoke tests |
blueorangutan
commented
Nov 18, 2020
Trillian test result (tid-3197)
|
yadvr
commented
Nov 18, 2020
Needs a round of test against vmware |
blueorangutan
commented
Nov 18, 2020
@rhtyd a Trillian-Jenkins test job (centos7 mgmt + vmware-67u3) has been kicked to run smoke tests |
blueorangutan
commented
Nov 19, 2020
Trillian test result (tid-3207)
|
DaanHoogland
commented
Nov 20, 2020
I do not see how these template tests (all pertaining to checksums are related. We may have uncovered an unrelated issue thought it does not happen in other vmware tests. @rhtyd , you want to make the call? |
yadvr
commented
Nov 20, 2020
Agree, seems like env related |
Description
When vpc tiers are added/removed and public ips in different range is added/removed, the nics in VPC VRs are added/removed, it leads to inconsecutive device in VPC VRs.
When we reboot the VR, (or VR is rebooted by some other reasons like HA), the ips in VPC VRs might be assigned to wrong interfaces.
Types of changes
Screenshots (if appropriate):
How Has This Been Tested?