OK, I'm sure I'll get flamed off the list for this, but after the long discussion about RHEL6 features which spiraled into conversations about minimal installs, where Redhat seems to constantly fight what most of us seem to want, I started thinking about other things which cause me grief regarding RHEL and the reality is, the most common annoyance for me is Redhat's dismissive attitude toward regressions.
For example, we've been Redhat customers since late 2003 and currently run about 25 RHEL servers. Certainly we're not a huge customer, but a good paying customer. In that time I've opened 14 support tickets and, of those, 8 were caused by regressions introduced during an update of the current major release (for example when moving from RHEL4 U2 to RHEL4 U3). If I include tickets that were opened due to regressions when upgrading between major releases (upgrade from RHEL3 to RHEL4, or RHEL4 to RHEL5) then it's actually 12 of the 14 support tickets that were caused by regressions from previous versions. That means 85% of my tickets were caused by Redhat breaking something that was previously working. When I open these tickets they are not treated with any priority, I'm usually told stuff like, "we're aware of the issue and it will be fixed in a future update". Typically it has taken between 6-9 months to get those regressions fixed, and in many cases much longer. I fail to understand how a company that wants to be taken seriously in the enterprise space can be so poorly focused on the impact of regressions, especially within a major release. If I were making my wish list for RHEL6 it would be that the product would have a much higher focus on minimizing regressions, and, when they do occur, a focus on informing customers about them, and getting them corrected quickly and with priority. Am I the only person who feels this way or has this problem? Have I just been the most unlucky admin using Redhat by being hit with so many regressions? I don't think that's the case because I've seen people hit with regressions that didn't impact me (nss_ldap, random ethernet device order, this list could go on) but I'd love to hear others thoughts. Thanks, Tom _______________________________________________ rhelv5-list mailing list [email protected] https://www.redhat.com/mailman/listinfo/rhelv5-list
