OESF | ELSI | pdaXrom | OpenZaurus | Zaurus Themes | Community Links | Ibiblio

IPB

Welcome Guest ( Log In | Register )

 
Reply to this topicStart new topic
> Too Much Uptime, when /proc/uptime wraps
cvmiller
post Oct 3 2005, 05:30 PM
Post #1





Group: Members
Posts: 242
Joined: 31-March 04
Member No.: 2,592



Hey All,

I have two Zaurus's, a SL-6000 which I use as my PDA everywhere, and my old SL-5500, which spends most of its life on a shelf, turned on, and on my wireless network (a great thing to ping when I need to test the wireless network).

So the SL-5500 has lots of uptime, since it is on 24x7. I have been monitoring the uptime, and it has been incrementing nicely. I thought it was doing pretty well after it passed 400 days of uptime.

Imagine my surprise when I checked the uptime today to see that it is only 7 days of uptime! Checking the last log, it is clear that the SL-5500 hasn't rebooted since 9 April 2004, so something was off.

Not the end of the world. I guess Sharp (or Embedix) wasn't planning on the Zaurus being "up" for so long. The little Linux box that could.

Craig...
Go to the top of the page
 
+Quote Post
inode0
post Oct 3 2005, 05:50 PM
Post #2





Group: Members
Posts: 97
Joined: 15-December 04
Member No.: 5,933



QUOTE(cvmiller @ Oct 4 2005, 01:30 AM)
Not the end of the world. I guess Sharp (or Embedix) wasn't planning on the Zaurus being "up" for so long. The little Linux box that could.

I wouldn't blame them ... might it be the old 497 day bug?! My woody 2.4.18 kernel rolled its counter over after 497 days too.

John
Go to the top of the page
 
+Quote Post
Optogeek
post Oct 4 2005, 06:25 AM
Post #3





Group: Members
Posts: 48
Joined: 3-February 04
From: Edinburgh, Scotland
Member No.: 1,681



This is from memory so I may be wrong....

The internal uptime clock in 2.4 linux kernels is 100Hz (not MHz) so with a 32bit counter you can only measure 497 days before you get wrap around. On new kernels the internal clock is 1000Hz and you get a wrap around after 49.7 days.
Go to the top of the page
 
+Quote Post
cvmiller
post Oct 5 2005, 05:52 PM
Post #4





Group: Members
Posts: 242
Joined: 31-March 04
Member No.: 2,592



QUOTE(inode0 @ Oct 3 2005, 05:50 PM)
I wouldn't blame them ... might it be the old 497 day bug?! My woody 2.4.18 kernel rolled its counter over after 497 days too.

John
*


OK, I guess that make sense. I have a 2.6 machine, I'll have to see if it lasts longer than 45 days (I read that the /proc/uptime counter was increased to 64 bits in 2.6).

Thanks,

Craig...
Go to the top of the page
 
+Quote Post
inode0
post Oct 5 2005, 05:56 PM
Post #5





Group: Members
Posts: 97
Joined: 15-December 04
Member No.: 5,933



QUOTE(cvmiller @ Oct 6 2005, 01:52 AM)
OK, I guess that make sense. I have a 2.6 machine, I'll have to see if it lasts longer than 45 days (I read that the /proc/uptime counter was increased to 64 bits in 2.6).


It will last a lot longer than 45 days. smile.gif

John
Go to the top of the page
 
+Quote Post

Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 22nd May 2013 - 02:51 AM