You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@trafficserver.apache.org by zwoop <gi...@git.apache.org> on 2016/03/08 22:15:42 UTC
[GitHub] trafficserver pull request: TS-4207 Adds better checks to avoid Ho...
GitHub user zwoop opened a pull request:
https://github.com/apache/trafficserver/pull/521
TS-4207 Adds better checks to avoid HostDB crashes in 6.x
As much as I dislike this bandaid fix, it does prevent the crashes that we've
been experiencing since upgrading to 6.1.x. It's also in line with other
checks for the validity of the rr done in other places. Since the crash is
rather infrequent (once or twice a day on ~30 machines), it's difficult to
track down the exact commit that broke this.
But I did run a build that reverted all HostDB changes since 5.3.x and that
also fixes the crasher. So it's definitely either a fix triggering an old bug,
or a regression.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/zwoop/trafficserver TS-4207
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/trafficserver/pull/521.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #521
----
commit 19b233306b89e5e9ad32dc69a957f90f4312b306
Author: Leif Hedstrom <zw...@apache.org>
Date: 2016-03-07T22:34:56Z
TS-4207 Adds better checks to avoid HostDB crashes in 6.x
As much as I dislike this bandaid fix, it does prevent the crashes that we've
been experiencing since upgrading to 6.1.x. It's also in line with other
checks for the validity of the rr done in other places. Since the crash is
rather infrequent (once or twice a day on ~30 machines), it's difficult to
track down the exact commit that broke this.
But I did run a build that reverted all HostDB changes since 5.3.x and that
also fixes the crasher. So it's definitely either a fix triggering an old bug,
or a regression.
----
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] trafficserver pull request: TS-4207 Adds better checks to avoid Ho...
Posted by zwoop <gi...@git.apache.org>.
Github user zwoop closed the pull request at:
https://github.com/apache/trafficserver/pull/521
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] trafficserver pull request: TS-4207 Adds better checks to avoid Ho...
Posted by jacksontj <gi...@git.apache.org>.
Github user jacksontj commented on the pull request:
https://github.com/apache/trafficserver/pull/521#issuecomment-195071103
I don't think this should land. If you look above there is already a check if the item is round robin (`if (r->round_robin) {`) so this check would basically just be checking if the item is malformed, not actually doing a real check. If this does stop your crash from happening, that means there is an entry in hostdb that has `round_robin` and no `rr`-- so that does narrow down the problem quite a bit.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] trafficserver pull request: TS-4207 Adds better checks to avoid Ho...
Posted by zwoop <gi...@git.apache.org>.
Github user zwoop commented on the pull request:
https://github.com/apache/trafficserver/pull/521#issuecomment-195044366
Should I land this? I understand it might hide an underlying problem, but a) it does work and b) it does make sense to check the rr, we do that in other places too.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] trafficserver pull request: TS-4207 Adds better checks to avoid Ho...
Posted by zwoop <gi...@git.apache.org>.
Github user zwoop commented on the pull request:
https://github.com/apache/trafficserver/pull/521#issuecomment-195107210
Ok. So we'll leave 6.x unusable for now. :). I understand you primary concern, that it hides possibly another problem. But your r->round_robin argument doesn't make sense, if you look above there, we do exactly what my patch does already in the if() above. Why did we check for r->rr in there already?
But I'll close this, and we leave 6.x unusable for now.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---