Cheer up statsbot!

*makes predictable and lame 'going down' joke*

makes predictable and lame 'going down' joke

yes? go on.

The machine seems ok:

uptime

21:36:04 up 407 days, 16:33, load average: 0.00, 0.00, 0.00

df

Filesystem Size Used Available Use% Mounted on
/dev/mtdblock1 30.8M 10.1M 20.6M 33% /
/dev/mmcblk0p1 967.3M 31.7M 935.7M 3% /home

But the statsbot process keeps terminating.

nothing on /var or /tmp?

pipe breaking? can't bind to port?

strace it?

Not any more. I just rebooted it.

On the upside, I now know that the root prompt makes big fonts!

ha!

what, you mean you don't just have a robot clicking refresh on the who's online page repeatedly and writing the answers down with pen and paper? i know a robot could just remember it, but writing it down makes it more official and scientific.

Oh crap, I forgot about the ntpd bug. Statsbot thinks it's 1970 again.

# date
Thu Jan 1 11:02:38 AEDT 1970
#

Gratuitous picture of robot with afro 'do.

yowser// tech talk.//I'm a running like in that goodies episode

.. or not.

ntpd is going now. I really wasn't looking forward to wide lapels again.

# date
Mon Jul 13 21:51:54 AEST 2009
#

It's there in the process list...

# ps  
PID  Uid     VmSize Stat Command  
...  
340 root        556 S   -sh  
360 root        540 S   ntpd -g  
384 root       1288 S   microperl statsbot.pl  
387 root        364 R   ps  
#  

would running it with -w helped?

Perl -w? Don't think so; I'm fairly sure there are no bugs that would be picked up by -w. Also, microperl probably doesn't implement it.

We just have to wait until 22:30 to know whether it's going to work.

The 22:30 sample worked. Now we have to wait for the graph update at 22:45.

It works! Statsbot is now up!

Is HAL 9000 angry and threatening to cut your internet access AGAIN?

...how about sending Robocop's ED-209 enforcement droid in to their office? With robot brotherhood (I saw that episode of Lost In Space, I know they help each other out) he could ensure 'reconfiguration' disruption is minimal, or reconfigure some nerd bacon.

Statsbot was down from 5pm yesterday to 9am today. I'm not sure why. The little fellow has been rebooted and now seems to be collecting stats as usual.

reconfigure some nerd bacon

I don't know what that means, but I like the sound of it.

i assume it was a sopa blackout.

Down again. I think it might be sick.

Statsbot will be down this weekend due to a planned power outage at the Statsbot World Domination HQ. The power outage (love that word) is scheduled to start at noon Saturday.

Statsbot won't be back up until Monday.

I hope you'd getting time and a half, statsbot.

Dunno what's going on here, but statsbot's not been talking to the world since yesterday morning. I think a reboot might be in order.

BUG: soft lockup detected on CPU#0!
[] (dump_stack+0x0/0x14) from [] (softlockup_tick+0x9c/0xbc)
[] (softlockup_tick+0x0/0xbc) from [] (run_local_timers+0x18/0x1c)
r7 = 0000001A r6 = 00000000 r5 = 00000000 r4 = C0367040
[] (run_local_timers+0x0/0x1c) from [] (update_process_times+0x44/0x70)
[] (update_process_times+0x0/0x70) from [] (timer_tick+0xc4/0xe4)
r5 = C01F70F0 r4 = C01F1688
[] (timer_tick+0x0/0xe4) from [] (pxa_timer_interrupt+0x48/0x9c)
r5 = 00000000 r4 = C01E5B9C
[] (pxa_timer_interrupt+0x0/0x9c) from [] (handle_IRQ_event+0x9c/0xe4)
[] (handle_IRQ_event+0x0/0xe4) from [] (handle_level_irq+0xb0/0x108)
r7 = 0000015A r6 = C5FC5EC0 r5 = 0000001A r4 = C01E0680
[] (handle_level_irq+0x0/0x108) from [] (asm_do_IRQ+0x44/0x60)
r5 = C01E0680 r4 = 0000001A
[] (asm_do_IRQ+0x0/0x60) from [] (__irq_svc+0x30/0x80)
r6 = 04000000 r5 = C5FC5DDC r4 = FFFFFFFF
[] (smc911x_interrupt+0x0/0x758 [smc911x]) from [] (handle_IRQ_event+0x9c/0xe4)
[] (handle_IRQ_event+0x0/0xe4) from [] (handle_edge_irq+0x10c/0x150)
r7 = C5FC5F34 r6 = C75B4920 r5 = 00000083 r4 = C01E20C0
[] (handle_edge_irq+0x0/0x150) from [] (pxa_gpio_demux_handler+0x128/0x158)
r6 = 00000084 r5 = C01E20C0 r4 = 00000001
[] (pxa_gpio_demux_handler+0x0/0x158) from [] (asm_do_IRQ+0x44/0x60)
r6 = C5FC5FB0 r5 = C01E0280 r4 = 0000000A
[] (asm_do_IRQ+0x0/0x60) from [] (__irq_svc+0x30/0x80)
r6 = 00000400 r5 = C5FC5EF4 r4 = FFFFFFFF
[] (process_backlog+0x0/0x17c) from [] (net_rx_action+0x84/0x160)
[] (net_rx_action+0x0/0x160) from [] (__do_softirq+0x60/0xd4)
r8 = 0000A210 r7 = BEB73D74 r6 = 00000008 r5 = C01F6F14
r4 = 00000001
[] (__do_softirq+0x0/0xd4) from [] (irq_exit+0x44/0x4c)
r6 = 00000000 r5 = C01E0280 r4 = 0000000A
[] (irq_exit+0x0/0x4c) from [] (asm_do_IRQ+0x48/0x60)
[] (asm_do_IRQ+0x0/0x60) from [] (__irq_usr+0x40/0x80)
r6 = 00000400 r5 = 00000000 r4 = FFFFFFFF

Hmmm, maybe if you softcockup in the dump stack backlog that might give your handle edge the asm it needs.

Please leave my soft cock out of this, Kez.

Glad I could help.

Oh dear. Statsbot seems to be down again. The last update was 10:45 on the 4th of January. Probably the heat or something. Poor little dear.

I'll give him a stern talking to when I get back on Monday.

Who read the stat, statbot?

What was it like the first day we went off the chart? What was the controversy that caused it?

Statsbot will be down this weekend. It's moving from the office to Mrs. Bot's house in the 'burbs.

It's back up.
Statsbot is now hiding in a corner behind the tv.

Have the stats hit the decade yet?

It'll be ten years in another few months.

Collate some interesting data for us when it clocks over, please.