Comments for domsch.com blog http://domsch.com/blog "It Just Works" is more than just a slogan, it's a way of life! Wed, 27 Jun 2012 12:21:35 +0000 hourly 1 http://wordpress.org/?v=3.5.1 Comment on logrotate and bash by Resuna http://domsch.com/blog/?p=592&cpage=1#comment-1987 Resuna Wed, 27 Jun 2012 12:21:35 +0000 http://domsch.com/blog/?p=592#comment-1987 Aha! I didn’t know about that option. Copytruncate should work in most cases, because redirection onto the end of a file opens the file in “append” mode, so it seeks to the end on every write.

]]> Comment on logrotate and bash by adobriyan http://domsch.com/blog/?p=592&cpage=1#comment-1983 adobriyan Wed, 27 Jun 2012 09:53:44 +0000 http://domsch.com/blog/?p=592#comment-1983 See “copytruncate”.

]]> Comment on logrotate and bash by Resuna http://domsch.com/blog/?p=592&cpage=1#comment-1966 Resuna Tue, 26 Jun 2012 11:28:26 +0000 http://domsch.com/blog/?p=592#comment-1966 As I was reading through this, I was thinking that UNIX file descriptors didn’t work that way… the previous commenter isn’t quite correct as to how it works: it’s not that the find gets a complete copy of the file descriptor (it doesn’t, it shares a seek position with bash), but once bash closes that fd it doesn’t matter what it does past that point, but the fact remains that it doesn’t work.

The correct way to handle logging is to write to a pipe to a program that’s logrotate-aware, or to write to a program that feeds lines to syslog. There are a variety of such programs, I seem to recall there being one included in qmail or tcptools.

]]> Comment on logrotate and bash by syskill http://domsch.com/blog/?p=592&cpage=1#comment-1962 syskill Mon, 25 Jun 2012 20:13:54 +0000 http://domsch.com/blog/?p=592#comment-1962 I’ve tested this and it doesn’t work. :-( The find process inherits *copies* of the bash process’ stdout and stderr file descriptors as it is spawned. When the bash process reopens its file descriptors, that has no bearing on the find process.

]]> Comment on logrotate and bash by Joe Klemmer http://domsch.com/blog/?p=592&cpage=1#comment-1960 Joe Klemmer Mon, 25 Jun 2012 18:47:14 +0000 http://domsch.com/blog/?p=592#comment-1960 Nice and relatively simple. Thanks for sharing.

]]> Comment on Dell Linux Engineers work over 5000 bugs with Red Hat by Links for March 30th through April 4th — Vinny Carpenter's blog http://domsch.com/blog/?p=589&cpage=1#comment-1919 Links for March 30th through April 4th — Vinny Carpenter's blog Fri, 06 Apr 2012 01:24:26 +0000 http://domsch.com/blog/?p=589#comment-1919 [...] Dell Linux Engineers work over 5000 bugs with Red Hat | domsch.com blog – This is a testament to the hard work Dell puts into ensuring Linux "Just Works" on our servers, straight out of the box, with few to no extra driver disks or post-install updates needed to make your server fully functional. [...]

]]> Comment on Dell Linux Engineers work over 5000 bugs with Red Hat by Bob Plankers http://domsch.com/blog/?p=589&cpage=1#comment-1914 Bob Plankers Wed, 04 Apr 2012 14:37:42 +0000 http://domsch.com/blog/?p=589#comment-1914 Matt, that’s very cool, and shows more dedication to Red Hat Enterprise Linux than even Red Hat has. I say that because Red Hat has never worked on or fixed a bug I’ve reported to them via their own paid support channels. Reporting it via Dell (or VMware) has always meant it will get looked at. They likely owe much of their customer base to you guys, and I’m hoping that relationship continues.

]]> Comment on Dell Linux Engineers work over 5000 bugs with Red Hat by Charlie http://domsch.com/blog/?p=589&cpage=1#comment-1912 Charlie Tue, 03 Apr 2012 15:51:03 +0000 http://domsch.com/blog/?p=589#comment-1912 Well, Matt, that’s admittedly a Good Thing [tm], but every time I boot Red Hat Enterprise Linux v5 on a Dell Blade I still see this:

Call Trace:
[] setup_irq+0x1b7/0x1cf
[] serial8250_interrupt+0×0/0xfe
[] request_irq+0xb0/0xd6
[] serial8250_startup+0x43d/0x5dc
[] uart_startup+0×76/0x16c
[] uart_open+0x19e/0×427
[] tty_open+0x1e8/0x3b0
[] chrdev_open+0x14d/0×183
[] chrdev_open+0×0/0×183
[] __dentry_open+0xd9/0x1dc
[] do_filp_open+0x2a/0×38
[] do_sys_open+0×44/0xbe
[] tracesys+0xd5/0xe0

So there’s still lots of room for improvement.

]]> Comment on Dell Linux Engineers work over 5000 bugs with Red Hat by Tracy Reed http://domsch.com/blog/?p=589&cpage=1#comment-1903 Tracy Reed Mon, 02 Apr 2012 07:23:05 +0000 http://domsch.com/blog/?p=589#comment-1903 mdomsch, please check this out:

https://lopsa.org/pipermail/tech/2011-January/005620.html

I wrote it a little over a year ago. Since then I have had plenty of other “fun” times with Dell and Linux. Like the R415′s I just deployed which would not work out of the box attached to a Dell 24 drive 2.5″ SAS JBOD via the external SAS connector (I don’t have the JBOD model number handy at the moment). I forget the particule error which was given which prevented the machine from booting but a RAID card firmware upgrade fixed it. They changed the install script in the firmware upgrade so I couldn’t just replace RHEL with CentOS or some such also. So I had to boot a CD ROM image. If it isn’t one thing it’s another…

]]> Comment on Dell Linux Engineers work over 5000 bugs with Red Hat by Rex http://domsch.com/blog/?p=589&cpage=1#comment-1900 Rex Sun, 01 Apr 2012 19:08:49 +0000 http://domsch.com/blog/?p=589#comment-1900 … but I still cannot find a reasonably configured desktop or laptop computer from Dell retail channel that has any flavor of Linux pre-installed. If I want a laptop, I have to pay the Microsoft tax. — Just plain sux.

]]>