Bug 1598491
| Summary: | RFE: HAProxy does not resolve IPv6 resolvable hostnames in the backend section | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Product: | Red Hat Enterprise Linux 7 | Reporter: | Robert Scheck <redhat-bugzilla> | ||||||
| Component: | haproxy | Assignee: | Ryan O'Hara <rohara> | ||||||
| Status: | CLOSED ERRATA | QA Contact: | Brandon Perkins <bperkins> | ||||||
| Severity: | medium | Docs Contact: | |||||||
| Priority: | unspecified | ||||||||
| Version: | 7.5 | CC: | aherr, cfeist, pbarta, redhat-bugzilla, robert.scheck, rohara, sababu | ||||||
| Target Milestone: | rc | Keywords: | FutureFeature | ||||||
| Target Release: | 7.6 | ||||||||
| Hardware: | All | ||||||||
| OS: | Linux | ||||||||
| Whiteboard: | |||||||||
| Fixed In Version: | haproxy-1.5.18-8.el7 | Doc Type: | Bug Fix | ||||||
| Doc Text: |
Cause: HAProxy was not using getaddrinfo() to resolve IPv6 hostnames.
Consequence: When a backend server was referred to by hostname and resovled to an IPv6 address, HAProxy would alert the user that the server had an invalid address. This would result in the server not being used.
Fix: Compile HAProxy with th USE_GETADDRINFO flag, which will cause HAProxy to use getaddrinfo() to resolve hostnames instead of gethostbyname().
Result: HAProxy will now resolve all valid hostnames at runtime.
|
Story Points: | --- | ||||||
| Clone Of: | Environment: | ||||||||
| Last Closed: | 2018-10-30 11:49:09 UTC | Type: | Bug | ||||||
| Regression: | --- | Mount Type: | --- | ||||||
| Documentation: | --- | CRM: | |||||||
| Verified Versions: | Category: | --- | |||||||
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||||
| Cloudforms Team: | --- | Target Upstream Version: | |||||||
| Embargoed: | |||||||||
| Attachments: |
|
||||||||
|
Description
Robert Scheck
2018-07-05 15:49:42 UTC
Created attachment 1456794 [details]
Patch suggestion for haproxy.spec
Cross-filed case 02135416 in the Red Hat customer portal. Ryan, what would you like to know? Hello Robert, I've updated the info about requested data in the case which is linked to this BZ, I believe you should have it in your email now. Anyway, repasting here, as it seems like you are monitoring the BZ directly as well: ------------------------------------- I will need config files and output from haproxy showing what does/does not work. While I agree that we should be compiling with USE_GETADDRINFO and it is a fix I put into Fedora, we should not compare haproxy-1.5 to haproxy-1.8 and they are very different in many regards, including some features that resolve backend sevrer hostnames at runtime, etc. We need output and config files so that we are sure we are fixing the core issue here (ie. something we can test). ------------------------------------- I understand it that complete haproxy config will be needed, steps to reproduce the issue, and log from haproxy covering the time when the issue is reproduced. It would recommend to set the haproxy to debug mode for the time of the issue reproducer, to have more verbose info. BR, Petr Petr, thanks. Just saw the GSS WoC after I pressed "Save Changes". Already working to get the requested information. (In reply to Robert Scheck from comment #10) > Ryan, what would you like to know? Steps to reproduce. In this case, a config file and output from haproxy showing any errors, etc. With that we have something to test/verify the the rebuild with USE_GETADDRINFO does indeed solve the problem. I'm confident it will fix the issue, but QA needs to test. Created attachment 1467560 [details]
Script with steps to reproduce the issue
Please let me know in case the steps are not good enough for QA or
if further information etc. are required.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2018:3317 |