Skip to content
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

subset lb doesn't properly remove subsets when being destroyed #3916

Closed
rgs1 opened this issue Jul 20, 2018 · 0 comments
Closed

subset lb doesn't properly remove subsets when being destroyed #3916

rgs1 opened this issue Jul 20, 2018 · 0 comments
Labels
Milestone

Comments

@rgs1
Copy link
Member

rgs1 commented Jul 20, 2018

We use hot restarts for our deploys. Everything works nicely with the subset lb, except for the fact the lb_subsets_active gauge is wrong after the old process exits. That is, the number of active subsets in the old process doesn't get decremented. From a quick look at the code:

https://github.com/envoyproxy/envoy/blob/master/source/common/upstream/subset_lb.cc#L57

there's nothing going on in SubsetLoadBalancer::~SubsetLoadBalancer(), where we could probably do the decrements.

Any concerns with doing the decrements from the destructor? Is there a better – or more standard perhaps — way for dealing with gauges when in the context of hot restarts?

cc: @zuercher @mattklein123

rgs1 pushed a commit to rgs1/envoy that referenced this issue Jul 20, 2018
Without these updates, the gauges will be incorrect after a hot
restart.

Fixes envoyproxy#3916

Signed-off-by: Raul Gutierrez Segales <rgs@pinterest.com>
@mattklein123 mattklein123 added this to the 1.8.0 milestone Jul 20, 2018
mattklein123 pushed a commit that referenced this issue Jul 20, 2018
Without these updates, the gauges will be incorrect after a hot
restart.

Fixes #3916
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

No branches or pull requests

2 participants