Bug 765598
| Summary: | Kickstart netinstall gives up instantly if download times-out | ||||||
|---|---|---|---|---|---|---|---|
| Product: | Red Hat Enterprise Linux 6 | Reporter: | Floris <bos> | ||||
| Component: | python-urlgrabber | Assignee: | James Antill <james.antill> | ||||
| Status: | CLOSED WONTFIX | QA Contact: | BaseOS QE - Apps <qe-baseos-apps> | ||||
| Severity: | medium | Docs Contact: | |||||
| Priority: | unspecified | ||||||
| Version: | 6.1 | CC: | zpavlas | ||||
| Target Milestone: | rc | ||||||
| Target Release: | --- | ||||||
| Hardware: | All | ||||||
| OS: | Linux | ||||||
| Whiteboard: | |||||||
| Fixed In Version: | Doc Type: | Bug Fix | |||||
| Doc Text: | Story Points: | --- | |||||
| Clone Of: | Environment: | ||||||
| Last Closed: | 2013-04-10 10:19:45 UTC | Type: | --- | ||||
| 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
Floris
2011-12-08 22:20:59 UTC
Created attachment 544205 [details]
log vt
I doubt there's anything that can be done. Based on your screenshot, your connection is close to unusable (less than 1 byte per second). Reassigning for completeness. Package installation in anaconda is handled by yum, and downloading is handled by the urlgrabber module. <<
your connection is close to unusable (less than 1 byte per second).
>>
No, the connection itself is fine.
However it is using a proxy server that waits until the entire file has been downloaded from the Internet completely, before passing the file to the client.
That's why it says less than 1 byte per second.
The proxy has not finished downloading the file from the Internet, so it has not started sending the file to the final destination either.
Such proxy servers are more common then you might think.
E.g. the ones that include decent virus scanning features, will need to download the entire file first to scan it, even though Yum only requests a small partial byte range of the file at the stage the error occurs.
This request was evaluated by Red Hat Product Management for inclusion in the current release of Red Hat Enterprise Linux. Because the affected component is not scheduled to be updated in the current release, Red Hat is unfortunately unable to address this request at this time. Red Hat invites you to ask your support representative to propose this request, if appropriate and relevant, in the next release of Red Hat Enterprise Linux. If you would like it considered as an exception in the current release, please ask your support representative. This issue is also present upstream, and not resolved yet. See http://lists.baseurl.org/pipermail/yum-devel/2012-June/009345.html Strictly speaking, it's NOT a urlgrabber bug. When a request fails, urlgrabber issues a failure_callback. The callback may return (in this case a retry is made), or re-raise the error (this skips remaining retries). When retrieving .rpm files, Yum always re-raises the error, skipping all retries. That's because using next mirror is almost always better than re-trying the same URL 10 times. Since by default each mirror is removed after failure and there's only 1 mirror, Yum makes 1 request only. The failover mechanism could be changed, but that should be done in Yum. I think this feature has to be implemented upstream first, and tuned on user input. If the lack of retries is a major problem in kickstarter, it can override the failure callback at any time with something like: def err_and_retry(errobj): print '%s: %s' % (errobj.url, errobj.exception) .. yb.repos.setFailureCallback(err_and_retry) |