Revert "Using map to store the listener in ConnectionHandlerImpl"#19811
Revert "Using map to store the listener in ConnectionHandlerImpl"#19811soulxu wants to merge 1 commit intoenvoyproxy:mainfrom
Conversation
|
@howardjohn let's revert it, looks like need some time to figure out what is happening here. thanks for helping on the test. |
lambdai
left a comment
lambdai
left a comment
There was a problem hiding this comment.
Thank you! I will triage after
Sorry for holding your further PRs on top of this one
|
@soulxu can you add DCO signed-off-by to the commit please? You'll have to force-push to make DCO-bot happy. |
…oyproxy#19362)" This reverts commit 67c2b72. Signed-off-by: He Jie Xu <hejie.xu@intel.com>
soulxu
force-pushed
the
revert-19362-using_map_for_conn_handler
branch
from
1de262f to
2d535ed
Compare
yea, just done that |
thanks for helping me here |
|
/wait I found out the root cause, working on a fix. |
| } | ||
|
|
||
| Address::InstanceConstSharedPtr Utility::getIpv4AnyAddress(uint32_t port) { | ||
| CONSTRUCT_ON_FIRST_USE(Address::InstanceConstSharedPtr, new Address::Ipv4Instance(port)); |
There was a problem hiding this comment.
I mess up with this. CONSTRUCT_ON_FIRST_USE will create static object for Ipv4Instance, so it is going to build the first time with the parameter passed in first time also.
There was a problem hiding this comment.
I found the same.
My local patch is to always return a newly created shared ptr. That will utilize the map lookup rather than iterating the map
There was a problem hiding this comment.
yea, that is.
I come up a fix #19802, but free to submit one if you want to since you already spend time on it.
Thanks for helping on this again!
|
Fix is here #19802, but free to merge this revert PR if this is better option. |
|
Close this PR, since we have a fix. |
3 participants
Reverts #19362
Since it is failling istio testsuits istio/istio#37004, getting the listener mix up. Before figure out what is happening here, let us revert the change.