Shining more light on the problem
Posted Aug 23, 2006 2:05 UTC (Wed) by gdt
In reply to: Shining more light on the problem
Parent article: Who maintains RPM?
I credit Red Hat for acknowledging the problem with a named individual...
Well I don't. This wasn't the first episode -- pressing Ctrl+C when running RPM once upon a time corrupted the database too. We were a paying customer and submitted a bug report. A similar level of lack of ownership and vitriol was displayed by the developer.
This was a showstopper bug for a production billing system. His supervisor didn't appear to take any action. There seemed to be no automated escalation. There seemed to be no level of management that would act upon our complaints and escalate the issue to a more capable developer. The local Red Hat office advised us that they had no influence. The resolution dragged on and on and was eventually to prevent Ctrl+C having any effect upon RPM. Of course that meant that any runaway RPM process could only be stopped with kill -9, which would of course corrupt the database.
I can't think of any other software company that would have allowed this sort of behaviour from an employee for such an extended period of time. It doesn't surprise me at all that other RPM bugs struck similar issues.
It is sad that so much effort was put into nagging at him rather than someone just fixing it and submitting a patch.
No. If we wanted that we'd use Debian. We pay Red Hat so that they take ownership of problems. If a Fedora user reports a showstopper problem we expect it to be addressed before it effects our applications running on RHEL. That's why we're happy to have our payments for RHEL used to subsidise the Fedora distribution.
Red Hat software maintenance is often more expensive than Solaris, AIX or Cisco -- and they include hardware maintenance too. We expect a level of service which matches the scale of those payments.
to post comments)