Bug 1349015
| Summary: | sshd_config should contain AcceptEnv TZ by default | ||
|---|---|---|---|
| Product: | Red Hat Enterprise Linux 7 | Reporter: | ervin.nemeth |
| Component: | openssh | Assignee: | Jakub Jelen <jjelen> |
| Status: | CLOSED WONTFIX | QA Contact: | BaseOS QE Security Team <qe-baseos-security> |
| Severity: | low | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | 7.4 | CC: | cponder, szidek |
| Target Milestone: | rc | ||
| Target Release: | --- | ||
| Hardware: | All | ||
| OS: | All | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | If docs needed, set a value | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2017-03-28 13:17:44 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: | |||
|
Description
ervin.nemeth
2016-06-22 13:55:07 UTC
> As I see, RedHat decided to whitelist the LANG and LC_* parameters with "AcceptEnv". Please consider adding TZ, too. The LANG and LC_* are send/accepted by the ssh/sshd based on the report in a bug #179851, 10 years ago. The arguments are mostly about the encoding, which might cause problems if not transferred to the remote session. But on the other hand requesting a language/locale/encoding that is not installed on the server causes problem in various tools unable to handle the errors. Both the above LANG and LC_* environment variables are standard in shell and set by default, unlike the TZ. Once you configure TZ in your local system (and you have got a reason to set this up to something different than system-wide value), you should be able to configure also the server with the same TZ or to send/accept this configuration. I don't think this should come by default. Thank you for the input and idea for improvement, but note that bugzilla is not a support tool. If you have a good reason and business requirement to see this in the next RHEL release, please get in touch with your Red Hat support: https://access.redhat.com/support/ I use a number of remote clusters with job-control systems like SLURM and PBS. When I look at the start-times and expected finish-times on running jobs, it's convenient to see them in my local time. I do travel a lot, so propagating the TZ variable from my laptop would be a very simple & seamless to deal with this. It's a pain to have to manually reset the TZ each time I log in, or manually edit my ~/.bash_profile each time I log on from a different timezone. I get that there would be a problem if the remote cluster didn't have a current list of timezones and couldn't recognize the one I'm sending. But none of this will happen if I don't use an explicit SendEnv in my ~/.ssh/config, anyway. So if I see problems on one of the remote clusters, I'll either not send the TZ to that one, or try to find a more canonical setting (like GMT-relative or something). Also, is this /etc/ssh/sshd_config inherited from the upstream Fedora development? |