#2201 Man pages do not specify that sssd dyndns_refresh_interval < 60 is pulled to 60 seconds
Closed: Fixed None Opened 5 years ago by jhrozek.

Ticket was cloned from Red Hat Bugzilla (product Red Hat Enterprise Linux 7): Bug 1054777

Description of problem:
If sssd AD or IPA provider has dyndns_refresh_interval specified as less than
60 seconds, actual refresh interval will be 60 seconds regardless.

This is because of the update rate limit hard-coded in ad_dyndns_update_send
and ipa_dyndns_update_send.

Version-Release number of selected component (if applicable):
python-sssdconfig-1.11.2-24.el7.noarch
sssd-krb5-1.11.2-24.el7.x86_64
sssd-krb5-common-1.11.2-24.el7.x86_64
sssd-tools-1.11.2-24.el7.x86_64
sssd-common-1.11.2-24.el7.x86_64
sssd-ad-1.11.2-24.el7.x86_64
sssd-1.11.2-24.el7.x86_64
sssd-client-1.11.2-24.el7.x86_64
sssd-ipa-1.11.2-24.el7.x86_64
sssd-proxy-1.11.2-24.el7.x86_64
sssd-common-pac-1.11.2-24.el7.x86_64
sssd-ldap-1.11.2-24.el7.x86_64

How reproducible:
Always

Steps to Reproduce:
1. Assign an address to a dummy interface.
2. Configure sssd for dynamic DNS updates either with AD or IPA provider.
3. Specify the dummy interface with dyndns_iface.
4. Specify 10 second refresh interval with dyndns_refresh_interval.
5. Start sssd.
6. Wait for initial dynamic DNS update.
7. Change the dummy interface address.
8. Wait 10 seconds.
9. Check if the DNS records correspond to the interface address.
10. Wait 50 seconds.
11. Check if the DNS records correspond to the interface address.

Actual results:
First check: DNS records are outdated.
Second check: DNS records are up-to-date.

Expected results:
First check: DNS records are up-to-date.
Second check: DNS records are up-to-date.

Additional info:
It would suffice to specify this limit in the documentation, but a better fix
would be making refresh interval less than 60 seconds effective, while
preserving the 60 seconds limit for longer intervals.

Fields changed

blockedby: =>
blocking: =>
changelog: =>
component: SSSD => Documentation
coverity: =>
design: =>
design_review: => 0
feature_milestone: =>
fedora_test_page: =>
milestone: NEEDS_TRIAGE => SSSD 1.13 beta
review: True => 0
selected: =>
testsupdated: => 0

Fields changed

mark: => 0

Fields changed

owner: somebody => preichl

Fields changed

milestone: SSSD 1.13 beta => SSSD 1.13 backlog
priority: major => trivial

Mass-moving tickets not planned for any immediate release and re-setting priority.

milestone: SSSD 1.13 backlog => SSSD Deferred
priority: trivial => major

Fields changed

keywords: => easyfix
owner: preichl => somebody
priority: major => trivial
sensitive: => 0
summary: Sssd dyndns_refresh_interval < 60 is pulled to 60 seconds => Man pages do not specify that sssd dyndns_refresh_interval < 60 is pulled to 60 seconds

I want to pick this up.

I hope these are steps which I need to follow
Owning the Ticket:
Modify Ticket ==> Blocked By: <my name=""> ==> Submit Changes

Making Code Changes, Submitting
I hope these are steps to change, test, submit:
1. Have latest sssd source code
# vim version.m4
# Primary version number
m4_define([VERSION_NUMBER], [1.14.90])
2. Reproduce the Issue. With steps mentioned in ticket.
3. Apply the Fix, test whether issue is fixed (locally). Also make intgchk
4. Submitting the patch:
# git add <path-tofile>
# git commit

Bugzilla and Fedora:
Since RHEL is fork of fedora, so do we need to own Bugzilla ticket also while owning the Fedora/sssd ticket.
Or We only change fedora/sssd ticket, code and it waterfalls to RHEL on upcoming release(if priotized)

Thanks

Replying to [comment:7 amitkumar25nov]:

I want to pick this up.

I hope these are steps which I need to follow
Owning the Ticket:
Modify Ticket ==> Blocked By: <my name=""> ==> Submit Changes

Blocked by is not the right field, you should see "Action" below "Change properties" and in action you should see "accept". If not, I need to grant you more permissions, but for now I'll manually assign the ticket for you.

Making Code Changes, Submitting
I hope these are steps to change, test, submit:
1. Have latest sssd source code
# vim version.m4
# Primary version number
m4_define([VERSION_NUMBER], [1.14.90])
2. Reproduce the Issue. With steps mentioned in ticket.
3. Apply the Fix, test whether issue is fixed (locally). Also make intgchk

In general yes, but since you will only be changing a man page, I think it's sufficient to run make and then check with "man ./path/to/the/buildroot/src/man/file" that the man page looks OK.

  1. Submitting the patch:
    # git add <path-tofile>
    # git commit

This is how you generate the commit locally. For review, we prefer github these days, please see https://fedorahosted.org/sssd/wiki/GithubWorkflow

Bugzilla and Fedora:
Since RHEL is fork of fedora, so do we need to own Bugzilla ticket also while owning the Fedora/sssd ticket.
Or We only change fedora/sssd ticket, code and it waterfalls to RHEL on upcoming release(if priotized)

We can handle bugzilla, feel free to just submit the PR

Fields changed

owner: somebody => amitkumar25nov

master:

milestone: SSSD Patches welcome => SSSD 1.15 Alpha
resolution: => fixed
status: new => closed

Metadata Update from @jhrozek:
- Issue assigned to amitkumar25nov
- Issue set to the milestone: SSSD 1.15.0

2 years ago

Login to comment on this ticket.

Metadata