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

[RFC] rgw: Libcurl+NSS memory leak mitigation when performing keystone SSL auth #20924

Closed
wants to merge 3 commits into from

Conversation

mkogan1
Copy link
Contributor

@mkogan1 mkogan1 commented Mar 15, 2018

Tracker issue:
http://tracker.ceph.com/issues/23375

When libcurl is configured with --with-nss
There is a memory leak in the PK11_CreateGenericObject() function (in libnss3.so)
that occurs when keystone users authority authentication is using ssl.
(rgw_keystone_verify_ssl = true)

By calling the curl_global_cleanup() its possible to release the memory in libcurl.

(This patch is currently only for the curl_easy_* flow, can be extended to the curl_multi_* flow also)

there is a memory leak in the libcurl+nss: PK11_CreateGenericObject() function,
rgw is affected when Keystone users authority authentication perfomed using ssl

Signed-off-by: Mark Kogan <mkogan@redhat.com>
if so enable the libcurl+nss memory cleaning option.

Signed-off-by: Mark Kogan <mkogan@redhat.com>
ceph.conf option to enable and configure the libcurl+nss memory cleaning interval.
(disabled by default)

Signed-off-by: Mark Kogan <mkogan@redhat.com>
@stale
Copy link

stale bot commented Oct 18, 2018

This pull request has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs.
If you are a maintainer or core committer, please follow-up on this issue to identify what steps should be taken by the author to move this proposed change forward.
If you are the author of this pull request, thank you for your proposed contribution. If you believe this change is still appropriate, please ensure that any feedback has been addressed and ask for a code review.

@stale stale bot added the stale label Oct 18, 2018
@mattbenjamin
Copy link
Contributor

@mkogan1 could this get a rebase?

@stale stale bot removed the stale label Dec 6, 2018
@mattbenjamin
Copy link
Contributor

@cbodley and I discussed--maybe the way forward here is to globally transition out of NSS crypto?

@mattbenjamin
Copy link
Contributor

@liewegas ^^

@liewegas
Copy link
Member

liewegas commented Dec 6, 2018

IIRC @rzarzynski said the backport is nontrivial, but it brings along some performance benefits with it. If it's feasible it seems like a better cost/benefit than spending time fixing old nss code.

@stale
Copy link

stale bot commented Feb 5, 2019

This pull request has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs.
If you are a maintainer or core committer, please follow-up on this issue to identify what steps should be taken by the author to move this proposed change forward.
If you are the author of this pull request, thank you for your proposed contribution. If you believe this change is still appropriate, please ensure that any feedback has been addressed and ask for a code review.

@stale stale bot added the stale label Feb 5, 2019
@stale
Copy link

stale bot commented May 6, 2019

This pull request has been automatically closed because there has been no activity for 90 days. Please feel free to reopen this pull request (or open a new one) if the proposed change is still appropriate. Thank you for your contribution!

@stale stale bot closed this May 6, 2019
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
3 participants