F13 boot lag

[GoinEasy9]

I'm just wondering if anyone else is seeing a lag during the boot. In another post I thought it was due to ABRT, because initially, when I disabled abrtd, the lag went away. Well, leaving abrtd disabled still didn't stop the lag during cold boot, although during a restart it seemed to help. Watching the boot scroll I can see it pausing both on Starting sendmail and Starting sm-client, each for about a minute. In the maillog it's saying sendmail: "My unqualified host name (Fedora13dw32) unknown: sleeping for retry, same with sm-msp-queue. My host naming conventions have been similar on F11 and F12. I also didn't have this problem until I updated to the 2.6.33-1 kernel. This is happening on both my 32 and 64 installs. I added a few apps since I installed for testing purposes, so I don't know if something else is affecting the boot.
Except for this 2 minute pause I found in messages:
Feb 27 18:17:13 Fedora13dw32 kernel: ACPI: I/O resource it87 [0x295-0x296] conflicts with ACPI region SIOE [0x290-0x2af]
Feb 27 18:17:13 Fedora13dw32 kernel: ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
Feb 27 18:19:14 Fedora13dw32 abrtd: Registered Analyzer plugin 'Kerneloops'
Feb 27 18:19:14 Fedora13dw32 abrtd: Registered Reporter plugin 'KerneloopsReporter'

I don't see anything else in the logs. Can anyone give me a clue as to where to look to resolve this?

[GoinEasy9]

So, I guess I'm the only one experiencing this. Guess I'll have to backtrack and see if something I installed is causing this.

[marvin_ita]

I found this bug open in bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=501596

And also this https://bugs.launchpad.net/fedora/+s...ux/+bug/382211

Seems to be related to lm-sensor... but don't know how to help you in fixing that.

Mattia

[GoinEasy9]

Seems to be related to lm-sensor

Thank You, When going through dmesg I remember seeing that bugzilla error. I also remember that sensors-detect changed the i2c module over to an it87 module, or something like that, I've run sensors-detect on that box for F11 and F12 and it always read chips and set up modules properly, so much so that I just accept the defaults and move on.
OK, so I start with lm_sensors. Thanks Again.

---------- Post added at 06:30 PM CST ---------- Previous post was at 02:03 PM CST ----------

Well, the setup of lm_sensors is the same in F13 as it was in F11 and F12, as are the modules used. So changes to lm_sensors and disabling abrtd didn't really do anything permanent.
Strange, today the first 2 cold boots didn't have lags, but a reboot (restart) had the lag, the opposite way it behaved yesterday.
You can see the lag in the maillog, the four lines at hour 14 did not have a lag, the six lines at hour 16 did have about a minute lag each when the boot scroll read Starting sendmail, and when it read Starting sm-client.

Mar 2 14:28:19 Fedora13dw32 sendmail[1371]: unable to qualify my own domain name (Fedora13dw32) -- using short name
Mar 2 14:28:19 Fedora13dw32 sendmail[2094]: starting daemon (8.14.4): SMTP+queueing@01:00:00
Mar 2 14:28:19 Fedora13dw32 sm-msp-queue[1379]: unable to qualify my own domain name (Fedora13dw32) -- using short name
Mar 2 14:28:19 Fedora13dw32 sm-msp-queue[2096]: starting daemon (8.14.4): queueing@01:00:00
Mar 2 16:29:35 Fedora13dw32 sendmail[1389]: My unqualified host name (Fedora13dw32) unknown; sleeping for retry
Mar 2 16:30:35 Fedora13dw32 sendmail[1389]: unable to qualify my own domain name (Fedora13dw32) -- using short name
Mar 2 16:30:35 Fedora13dw32 sendmail[1391]: starting daemon (8.14.4): SMTP+queueing@01:00:00
Mar 2 16:30:36 Fedora13dw32 sm-msp-queue[1398]: My unqualified host name (Fedora13dw32) unknown; sleeping for retry
Mar 2 16:31:36 Fedora13dw32 sm-msp-queue[1398]: unable to qualify my own domain name (Fedora13dw32) -- using short name
Mar 2 16:31:36 Fedora13dw32 sm-msp-queue[1400]: starting daemon (8.14.4): queueing@01:00:00

I found the spot in the code where the delay is taking place in /etc/rc.d/init.d/sendmail:

start() {
# Start daemons.
ret=0
updateconf
echo -n $"Starting $prog: "
daemon /usr/sbin/sendmail $([ "x$DAEMON" = xyes ] && echo -bd) \
$([ -n "$QUEUE" ] && echo -q$QUEUE) $SENDMAIL_OPTARG
RETVAL=$?
echo
[ $RETVAL -eq 0 ] && touch /var/lock/subsys/sendmail
let ret+=$RETVAL

if [ ! -f /var/run/sm-client.pid ]; then
echo -n $"Starting sm-client: "
touch /var/run/sm-client.pid
chown smmsp:smmsp /var/run/sm-client.pid
if [ -x /usr/sbin/selinuxenabled ] && /usr/sbin/selinuxenabled; then
/sbin/restorecon /var/run/sm-client.pid
fi
daemon --check sm-client /usr/sbin/sendmail -L sm-msp-queue -Ac \
-q$SMQUEUE $SENDMAIL_OPTARG
RETVAL=$?
echo
[ $RETVAL -eq 0 ] && touch /var/lock/subsys/sm-client
let ret+=$RETVAL
fi

[ $ret -eq 0 ] && return 0 || return 1
}

But I guess I'm not experienced enough to figure out which command is actually causing the delay. I see this as a learning experience so I might be thinking out loud here. If anyone sees something that I'm missing, I'd appreciate the help.

[CiaW]

I see a line in there about selinux, and as a knee-jerk response, I'm always willing to blame selinux.

 if [ -x /usr/sbin/selinuxenabled ] && /usr/sbin/selinuxenabled; then
/sbin/restorecon /var/run/sm-client.pid

Of course I could be completely wrong, and I don't use sendmail.

But when I saw that selinux line, I thought I'd throw out the suggestion to check the security or selinux log also. And are there any other log type messages (rsyslog?) showing up? Which just reminded me I have to install the gui for viewing log files, now to figure out what it's called...

---------- Post added at 06:52 PM CST ---------- Previous post was at 06:52 PM CST ----------

I see a line in there about selinux, and as a knee-jerk response, I'm always willing to blame selinux.

 if [ -x /usr/sbin/selinuxenabled ] && /usr/sbin/selinuxenabled; then
/sbin/restorecon /var/run/sm-client.pid

Of course I could be completely wrong, and I don't use sendmail.

But when I saw that selinux line, I thought I'd throw out the suggestion to check the security or selinux log also. And are there any other log type messages (rsyslog?) showing up? Which just reminded me I have to install the gui for viewing log files, now to figure out what it's called...

[GoinEasy9]

Thanks for your input CiaW
I have no need for selinux, so I disable it early, but, I too have looked at that line. FYI its gnome-system-log that your looking for. The secure log doesn't show any messages that happen during the minute time gap, so I'm assuming selinux is disabled. Sendmail is used by the system to send messages to the logs, and it seems to be working properly. So, I'm trying to figure out what causes this line:

Mar 2 16:29:35 Fedora13dw32 sendmail[1389]: My unqualified host name (Fedora13dw32) unknown; sleeping for retry

It appears when there are lags in the boot, and the lags are intermittent. Today when I started looking at it, the first two cold boots didn't have a lag, then when I did a restart, the boot lagged. Exactly opposite from yesterday, when the cold boots lagged and the restarts almost never lagged.
I'm also trying to remember a command/app I used a long time ago that gave me a vertical graph of the time sequence during boot. It showed each command during boot, and the time it took to complete each command and what processes and errors occurred during each command.
At least I'm looking at a lot of new things while trying to figure this out. Thanks Again.

---------- Post added 2010-03-03 at 12:27 AM CST ---------- Previous post was 2010-03-02 at 10:38 PM CST ----------

The app I was looking for was bootchart, but all it showed was that sendmail spent a long time during setup. Guess I'll keep looking.

[desertrat]

So - This might seem dumb, but is sendmail trying to connect to the internet for any reason?

During the last week to ten days I have noticed that about half of the internet would appear not to be available and my web browser (I use FireFox) and mailer would just sit and spin their wheels until they hit the time limits. The outgoing mail service from my provider still is not working, and I have to use another in order to send emails.

[GoinEasy9]

Thanks for the response desertrat, I don't see anything in the logs that indicate sendmail is trying to access the internet during boot. I've erased lm_sensors and that has had no effect on the intermittent lag. I still think the key is in the line:
Mar 2 16:29:35 Fedora13dw32 sendmail[1389]: My unqualified host name (Fedora13dw32) unknown; sleeping for retry
But my experience is lacking in the inner workings, and I still haven't figured a way to step through it in a fashion that would show me all the variables. Maybe I'll just live with it till the beta is released, then see if it persists after a re-install. Although I'd still like to know why it's happening.

[desertrat]

You might try renaming your computer.

Mine is something like Fedora.twelve I think

[GoinEasy9]

OMG, Have I just been spammed? I feel so dirty!!

(Edit: all gone now. You're a clean kitty once more! - Bob)

[GoinEasy9]

First of all THANKS Bob, I'm much better now.....

The lag during sendmail and sm-client got to me so I looked around some more. It's seems that others have had the same problem in the past:
http://linux.derkeiler.com/Mailing-L...5-02/0366.html
http://unix.derkeiler.com/Mailing-Li.../msg02184.html

Old posts, I know. But they helped. I thought of adding my host name to /etc/hosts, but didn't feel confident enough till I read those posts.

I SOLVED the problem by adding my host name to the end of the first line in /etc/hosts. I never had to do that before, and, after checking, the host name is not present there in either my F11 or F12 installs.

The messages in my maillog have changed also:
FROM:
Mar 8 21:16:22 Fedora14dw32 sendmail[1350]: My unqualified host name (Fedora14dw32) unknown; sleeping for retry
Mar 8 21:17:22 Fedora14dw32 sendmail[1350]: unable to qualify my own domain name (Fedora14dw32) -- using short name
Mar 8 21:17:22 Fedora14dw32 sendmail[1399]: starting daemon (8.14.4): SMTP+queueing@01:00:00
Mar 8 21:17:23 Fedora14dw32 sm-msp-queue[1406]: My unqualified host name (Fedora14dw32) unknown; sleeping for retry
Mar 8 21:18:23 Fedora14dw32 sm-msp-queue[1406]: unable to qualify my own domain name (Fedora14dw32) -- using short name
Mar 8 21:18:23 Fedora14dw32 sm-msp-queue[1408]: starting daemon (8.14.4): queueing@01:00:00

TO:
Mar 8 21:45:34 localhost sendmail[1320]: starting daemon (8.14.4): SMTP+queueing@01:00:00
Mar 8 21:45:34 localhost sm-msp-queue[1328]: starting daemon (8.14.4): queueing@01:00:00
Mar 8 21:45:37 localhost sendmail[1320]: restarting /usr/sbin/sendmail due to signal
Mar 8 21:45:37 localhost sm-msp-queue[1328]: restarting /usr/sbin/sendmail due to signal
Mar 8 21:45:37 localhost sendmail[1573]: starting daemon (8.14.4): SMTP+queueing@01:00:00
Mar 8 21:45:37 localhost sm-msp-queue[1577]: starting daemon (8.14.4): queueing@01:00:00

Well, I'm glad it's solved, but, I'd still like to know why. Especially since no one else chimed in with the same complaint.

[dd_wizard]

You may want to add it to two lines in /etc/hosts. Here's mine, and I've never touched it:

[gene@Mobile-PC ~]$ cat /etc/hosts
# hostname Mobile-PC added to /etc/hosts by anaconda
127.0.0.1   localhost localhost.localdomain localhost4 localhost4.localdomain4 Mobile-PC
::1         localhost localhost.localdomain localhost6 localhost6.localdomain6 Mobile-PC

dd_wizard

[GoinEasy9]

Funny, I've never touched any of mine and none have the host name I gave it included, and no others have the problem. They all look like this, from my F11 install:
127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4
::1 localhost localhost.localdomain localhost6 localhost6.localdomain6

I'll add it to the second line also, can't hurt.

Thanks dd_wizard