View previous topic :: View next topic |
Author |
Message |
carlos123 Guru
![Guru Guru](/images/ranks/rank_rect_3.gif)
![](images/avatars/gallery/TV Stars/picard1.jpg)
Joined: 12 Feb 2003 Posts: 536 Location: Alberta, Canada.
|
Posted: Tue Mar 18, 2003 5:10 am Post subject: Network services down or unable to restart peridically?? |
|
|
The last few days I have noticed that network related services seem to stop on their own for some strange reason. I can usually restart them once I notice that they are down but would appreciate any insight for why this might be happening from anyone.
Take my net.eth0 service. These days it's operating just fine but once in a while when I am away from my computer I come back to it and find it down for no apparent reason. So I have to restart with ....
/etc/init.d/net.eth0 stop
/etc/init.d/net.eth0 start
Then there is my ntpd daemon. It's even weirder with that one. Sometimes it goes down even when my net.eth0 service is up. And talk about weirdness. Check out the following command line attempts to get it working again...
Code: |
pine [~]
# /etc/init.d/ntpd stop
* Stopping ntpd...
* Failed to stop ntpd [ !! ]
pine [~]
# /etc/init.d/ntpd start
* WARNING: "ntpd" has already been started.
pine [~]
# /etc/init.d/ntpd restart
* Stopping ntpd...
* Failed to stop ntpd [ !! ]
pine [~]
# /etc/init.d/ntpd restart
* Stopping ntpd...
* Failed to stop ntpd [ !! ]
pine [~]
# /etc/init.d/ntpd stop
* Stopping ntpd...
* Failed to stop ntpd [ !! ]
pine [~]
# less /var/log/ntpd.log
pine [~]
# /etc/init.d/ntpd restart
* Stopping ntpd...
* Failed to stop ntpd [ !! ]
pine [~]
# /etc/init.d/ntpd start
* WARNING: "ntpd" has already been started.
pine [~]
# /etc/init.d/ntpd stop
* Stopping ntpd...
* Failed to stop ntpd [ !! ]
pine [~]
# /etc/init.d/ntpd start
* WARNING: "ntpd" has already been started.
pine [~]
|
It's not even showing up in a listing of services running. At least by the "ntpd" name! So how can it be already started when it doesn't even show up in a list of processes? Where would it be?
Anybody got any suggestions as to what might be causing these two network related services to stop like this or act weird? Out of the blue mostly when I am not at my computer for some reason? Could it be connected with my screen saver somehow?
Could it be that my computer just doesn't like me anymore .
Any ideas?
Thanks.
Carlos _________________ I'm not yet a real Guru so if you are a newbie here and want to teach me a thing or two, please do! I'm still learning just like you. |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
avenj Retired Dev
![Retired Dev Retired Dev](/images/ranks/rank-retired.gif)
![](images/avatars/96918131947dac2e645511.jpg)
Joined: 11 Oct 2002 Posts: 495 Location: New Hampshire
|
Posted: Tue Mar 18, 2003 5:37 am Post subject: |
|
|
Your second problem is because our init system keeps track of started services. /etc/init.d/<servicename> zap |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
Crg Guru
![Guru Guru](/images/ranks/rank_rect_3.gif)
![](images/avatars/6065622693e713182272fc.jpg)
Joined: 29 May 2002 Posts: 345 Location: London
|
Posted: Tue Mar 18, 2003 5:45 am Post subject: |
|
|
avenj is right for part 2. For part 1, what does dmesg tell you? What network card do you have? I'm asuuming from the questions you won't have changed mtu etc... |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
carlos123 Guru
![Guru Guru](/images/ranks/rank_rect_3.gif)
![](images/avatars/gallery/TV Stars/picard1.jpg)
Joined: 12 Feb 2003 Posts: 536 Location: Alberta, Canada.
|
Posted: Tue Mar 18, 2003 5:48 am Post subject: |
|
|
Thanks avenji. Just before I read your post, in an act of desperation, I finally got ntpd to start again by doing the zap thing.
Can you or some explain the sentence "Your second problem is because our init system keeps track of started services." as it relates to trying to stop a service and having Gentoo report back that it's already started? Even though it clearly isn't?
Does that mean that if a service is stopped abruptly by an unknown failure that init will still see it as started unless I zap it?
Does this mean that unless a service is stopped nicely and cleanly through the usual "/etc/init.d/<service> stop" routine that init will be thrown off kilter and may return weird messages such as a service already being started when there is no such service actually running?
Am I understanding that correctly? Can you or anyone else expand on what Gentoo does underneath with all this a bit more?
Whatever I learn I will put into the ntpd Documentation Tips and Tricks that I posted on ntpd.
As for these services stopping all of a sudden for no reason I guess I will just wait and see if I can catch them in the act. Although that might be difficult. They have NEVER stopped on me when I am in front of the computer. Only when I am away from it for an extended period of time - say at least an hour.
Thanks again.
Carlos _________________ I'm not yet a real Guru so if you are a newbie here and want to teach me a thing or two, please do! I'm still learning just like you. |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
avenj Retired Dev
![Retired Dev Retired Dev](/images/ranks/rank-retired.gif)
![](images/avatars/96918131947dac2e645511.jpg)
Joined: 11 Oct 2002 Posts: 495 Location: New Hampshire
|
Posted: Tue Mar 18, 2003 5:54 am Post subject: |
|
|
carlos123 wrote: | Thanks avenji. Just before I read your post, in an act of desperation, I finally got ntpd to start again by doing the zap thing.
Can you or some explain the sentence "Your second problem is because our init system keeps track of started services." as it relates to trying to stop a service and having Gentoo report back that it's already started? Even though it clearly isn't?
Does that mean that if a service is stopped abruptly by an unknown failure that init will still see it as started unless I zap it?
Does this mean that unless a service is stopped nicely and cleanly through the usual "/etc/init.d/<service> stop" routine that init will be thrown off kilter and may return weird messages such as a service already being started when there is no such service actually running?
|
Yes. To all of the above.
Quote: |
Am I understanding that correctly? Can you or anyone else expand on what Gentoo does underneath with all this a bit more?
Whatever I learn I will put into the ntpd Documentation Tips and Tricks that I posted on ntpd.
As for these services stopping all of a sudden for no reason I guess I will just wait and see if I can catch them in the act. Although that might be difficult. They have NEVER stopped on me when I am in front of the computer. Only when I am away from it for an extended period of time - say at least an hour.
Thanks again.
Carlos |
The short answer: dependencies.
Slightly longer answer: Most distributions use a standard sysvinit that doesn't do anything particularly complex, just starts and stops services in runlevels. We use a wrapper of sorts (most of the work is done by /sbin/rc) that provides for things like softlevels. Our init scripts can depend on other init scripts so that to start a certain service, another service needs to be started (and will be automagically), when one service is stopped any services depending on it are also stopped, and so on. This is very handy, but has downsides - you ran into one of them. In order to do this properly, the init system has to keep track of everything that's already been started. |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
carlos123 Guru
![Guru Guru](/images/ranks/rank_rect_3.gif)
![](images/avatars/gallery/TV Stars/picard1.jpg)
Joined: 12 Feb 2003 Posts: 536 Location: Alberta, Canada.
|
Posted: Tue Mar 18, 2003 5:56 am Post subject: |
|
|
I hope it's okay to post such a long message but here is the output of dmesg. If anybody sees anything weird I would appreciate hearing about it.
I have always thought it weird that Gentoo reports "ds: no socket drivers loaded!" but have never been able to see my way around that particular warning. I bolded that string in the following output for easy reference.
Quote: |
pine [~]
# dmesg
Linux version 2.4.19-gentoo-r10 (root@localhost) (gcc version 3.2.1 20021207 (Gentoo Linux 3.2.1-20021207)) #1 Thu Mar 6 08:50:42 MST 2003
BIOS-provided physical RAM map:
BIOS-e820: 0000000000000000 - 000000000009fc00 (usable)
BIOS-e820: 000000000009fc00 - 00000000000a0000 (reserved)
BIOS-e820: 00000000000f0000 - 0000000000100000 (reserved)
BIOS-e820: 0000000000100000 - 000000001fffc000 (usable)
BIOS-e820: 000000001fffc000 - 000000001ffff000 (ACPI data)
BIOS-e820: 000000001ffff000 - 0000000020000000 (ACPI NVS)
BIOS-e820: 00000000ffff0000 - 0000000100000000 (reserved)
511MB LOWMEM available.
Advanced speculative caching feature not present
On node 0 totalpages: 131068
zone(0): 4096 pages.
zone(1): 126972 pages.
zone(2): 0 pages.
Kernel command line: root=/dev/hdb2
Initializing CPU#0
Detected 451.028 MHz processor.
Console: colour VGA+ 80x25
Calibrating delay loop... 897.84 BogoMIPS
Memory: 513292k/524272k available (1051k kernel code, 8416k reserved, 409k data, 68k init, 0k highmem)
Dentry cache hash table entries: 65536 (order: 7, 524288 bytes)
Inode cache hash table entries: 32768 (order: 6, 262144 bytes)
Mount-cache hash table entries: 8192 (order: 4, 65536 bytes)
Buffer-cache hash table entries: 32768 (order: 5, 131072 bytes)
Page-cache hash table entries: 131072 (order: 7, 524288 bytes)
CPU: Before vendor init, caps: 0383f9ff 00000000 00000000, vendor = 0
CPU: L1 I cache: 16K, L1 D cache: 16K
CPU: L2 cache: 512K
CPU: After vendor init, caps: 0383f9ff 00000000 00000000 00000000
Intel machine check architecture supported.
Intel machine check reporting enabled on CPU#0.
CPU: After generic, caps: 0383f9ff 00000000 00000000 00000000
CPU: Common caps: 0383f9ff 00000000 00000000 00000000
CPU: Intel Pentium III (Katmai) stepping 03
Enabling fast FPU save and restore... done.
Enabling unmasked SIMD FPU exception support... done.
Checking 'hlt' instruction... OK.
POSIX conformance testing by UNIFIX
PCI: PCI BIOS revision 2.10 entry at 0xf08b0, last bus=1
PCI: Using configuration type 1
PCI: Probing PCI hardware
PCI: Using IRQ router PIIX [8086/7110] at 00:04.0
Limiting direct PCI/PCI transfers.
isapnp: Scanning for PnP cards...
isapnp: No Plug & Play device found
Linux NET4.0 for Linux 2.4
Based upon Swansea University Computer Society NET3.039
Initializing RT netlink socket
Starting kswapd
devfs: v1.12a (20020514) Richard Gooch (rgooch@atnf.csiro.au)
devfs: boot_options: 0x1
parport0: PC-style at 0x378 (0x778) [PCSPP(,...)]
parport0: irq 7 detected
pty: 256 Unix98 ptys configured
Serial driver version 5.05c (2001-07-0 with MANY_PORTS SHARE_IRQ SERIAL_PCI ISAPNP enabled
ttyS00 at 0x03f8 (irq = 4) is a 16550A
ttyS01 at 0x02f8 (irq = 3) is a 16550A
Uniform Multi-Platform E-IDE driver Revision: 6.31
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
PIIX4: IDE controller on PCI bus 00 dev 21
PIIX4: chipset revision 1
PIIX4: not 100% native mode: will probe irqs later
ide0: BM-DMA at 0xb800-0xb807, BIOS settings: hda:DMA, hdb:DMA
ide1: BM-DMA at 0xb808-0xb80f, BIOS settings: hdc:DMA, hdd:pio
hda: ST36422A, ATA DISK drive
hdb: ST310232A, ATA DISK drive
hdc: CRD-8240B, ATAPI CD/DVD-ROM drive
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
ide1 at 0x170-0x177,0x376 on irq 15
hda: 12500460 sectors (6400 MB) w/256KiB Cache, CHS=778/255/63, UDMA(33)
hdb: 20005650 sectors (10243 MB) w/512KiB Cache, CHS=1245/255/63, UDMA(33)
Partition check:
/dev/ide/host0/bus0/target0/lun0: p1 p2 p3 p4 < p5 p6 >
/dev/ide/host0/bus0/target1/lun0: p1 p2 p3 p4
Floppy drive(s): fd0 is 1.44M
FDC 0 is a post-1991 82077
Linux Kernel Card Services 3.1.22
options: [pci] [cardbus] [pm]
NET4: Linux TCP/IP 1.0 for NET4.0
IP Protocols: ICMP, UDP, TCP
IP: routing cache hash table of 4096 buckets, 32Kbytes
TCP: Hash tables configured (established 32768 bind 65536)
NET4: Unix domain sockets 1.0/SMP for Linux NET4.0.
ds: no socket drivers loaded!
reiserfs: checking transaction log (device 03:42) ...
Using r5 hash to sort names
ReiserFS version 3.6.25
VFS: Mounted root (reiserfs filesystem) readonly.
Mounted devfs on /dev
Freeing unused kernel memory: 68k freed
Adding Swap: 240964k swap-space (priority -1)
PCI: Found IRQ 11 for device 00:0a.0
3c59x: Donald Becker and others. www.scyld.com/network/vortex.html
00:0a.0: 3Com PCI 3c905C Tornado at 0xb000. Vers LK1.1.16
usb.c: registered new driver usbdevfs
usb.c: registered new driver hub
usb-uhci.c: $Revision: 1.275 $ time 09:11:44 Mar 6 2003
usb-uhci.c: High bandwidth mode enabled
PCI: Found IRQ 10 for device 00:04.2
usb-uhci.c: USB UHCI at I/O 0xb400, IRQ 10
usb-uhci.c: Detected 2 ports
usb.c: new USB bus registered, assigned bus number 1
hub.c: USB hub found
hub.c: 2 ports detected
usb-uhci.c: v1.275:USB Universal Host Controller Interface driver
devfs_mk_dir(input): using old entry in dir: c25100c0 ""
mice: PS/2 mouse device common for all mice
reiserfs: checking transaction log (device 03:43) ...
Using r5 hash to sort names
ReiserFS version 3.6.25
hub.c: USB new device connect on bus1/1, assigned device number 2
usb.c: USB device 2 (vend/prod 0x46d/0xc501) is not claimed by any active driver.
usb.c: registered new driver hid
input0,hiddev0: USB HID v1.10 Mouse [Logitech USB Receiver] on usb1:2.0
hid-core.c: v1.8.1 Andreas Gal, Vojtech Pavlik <vojtech@suse.cz>
hid-core.c: USB HID support drivers
hdc: ATAPI 24X CD-ROM drive, 128kB Cache, DMA
Uniform CD-ROM driver Revision: 3.12
SCSI subsystem driver Revision: 1.00
reiserfs: checking transaction log (device 03:06) ...
Using r5 hash to sort names
ReiserFS version 3.6.25
reiserfs: checking transaction log (device 03:06) ...
Using r5 hash to sort names
ReiserFS version 3.6.25
pine [~]
#
|
My network card is a 3Com 905C-TX-M Etherlink 10/100 PCI.
Thanks.
Carlos _________________ I'm not yet a real Guru so if you are a newbie here and want to teach me a thing or two, please do! I'm still learning just like you. |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
carlos123 Guru
![Guru Guru](/images/ranks/rank_rect_3.gif)
![](images/avatars/gallery/TV Stars/picard1.jpg)
Joined: 12 Feb 2003 Posts: 536 Location: Alberta, Canada.
|
Posted: Tue Mar 18, 2003 6:11 am Post subject: |
|
|
I just thought of something that might contribute to this weird stopping of services when I am away from the computer.
My on-system clock battery might be dead or dying on me. I think this might be the case because before I started using ntpd my clock was WAY off. And very erratic. I just never took the time to get a new battery and replace mine (bit of a hassle).
Now that I have been using ntpd it usually sets my clock just fine and keeps it accurate so now I have even more reason to ignore my clock battery state.
Could a dead clock battery somehow trigger a service stopper when ntpd does an update of some kind which in turn might also cause my net.eth0 service to stop as well? I have noticed that sometimes ntpd reports that because my clock was off by so much (after such a stoppage) that it refuses to start unless I bring the clock time up to a more or less accurate setting first. Before continuing my efforts to start it.
Could a clock glitch that happens in connection with my screen saver cause a shut down of ntpd and cascade into shutting down my network service too?
I know I am just fishing in the dark here but I thought I would pass that theory by you all in case there might be some water there.
Thanks.
Carlos
PS. Is there any pressing reason for me to change my clock battery . _________________ I'm not yet a real Guru so if you are a newbie here and want to teach me a thing or two, please do! I'm still learning just like you. |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
carlos123 Guru
![Guru Guru](/images/ranks/rank_rect_3.gif)
![](images/avatars/gallery/TV Stars/picard1.jpg)
Joined: 12 Feb 2003 Posts: 536 Location: Alberta, Canada.
|
Posted: Tue Mar 18, 2003 9:56 am Post subject: |
|
|
I just came back from an absence from my computer of about 2 or 3 hours and although the net.eth0 service was up and okay the ntpd service was down again.
Here are the last lines in the ntpd log file located at /var/log/ntpd.log...
Code: |
17 Mar 17:16:27 ntpd[4349]: signal_no_reset: signal 17 had flags 4000000
17 Mar 17:20:52 ntpd[4347]: time reset -0.179047 s
17 Mar 17:20:52 ntpd[4347]: kernel time discipline status change 41
17 Mar 17:20:52 ntpd[4347]: synchronisation lost
17 Mar 20:11:57 ntpd[4347]: synchronisation lost
17 Mar 20:20:34 ntpd[4347]: time reset 2.653502 s
17 Mar 20:20:34 ntpd[4347]: synchronisation lost
17 Mar 21:00:39 ntpd[4347]: ntpd exiting on signal 15
17 Mar 22:39:18 ntpd[5127]: frequency initialized 18.734 from /var/lib/misc/ntp.
drift
17 Mar 22:39:18 ntpd[5129]: signal_no_reset: signal 17 had flags 4000000
17 Mar 22:43:40 ntpd[5127]: kernel time discipline status change 41
18 Mar 00:27:20 ntpd[5127]: synchronisation lost
|
I don't know why synchronization would be lost so often given that I interogate no less than three different time servers listed in the /etc/ntp.conf file. Here is that section from the file...
Code: |
# PER CARLOS: added the following 3 lines.
server ntp.cpsc.ucalgary.ca
server timelord.uregina.ca
server ntp1.cs.wisc.edu
|
That is the ONLY change I made to the /etc/ntp.conf file. I left out the rest of the file from the listing.
Here is a listing of my /etc/conf.d/ntpd file...
Code: |
NTPDATE_CMD="ntpdate"
NTPDATE_OPTS="-b ntp.cpsc.ucalgary.ca"
|
Those are the ONLY changes made to the /etc/conf.d/ntpd configuration file. I left out the rest of the file and the comments between the two above lines.
When I run "ntpq -p" I get the error "ntpq: read: Connection refused".
The only descrepency that I can see is the fact that I do not list all three time servers in the NTPDATE_OPTS line of the /etc/conf.d/nptd file. Not sure how to do that but I guess I will try to just add the other two servers to that line with spaces in between and see if that helps unless anyone has any other suggestions.
Thanks.
Carlos _________________ I'm not yet a real Guru so if you are a newbie here and want to teach me a thing or two, please do! I'm still learning just like you. |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
carlos123 Guru
![Guru Guru](/images/ranks/rank_rect_3.gif)
![](images/avatars/gallery/TV Stars/picard1.jpg)
Joined: 12 Feb 2003 Posts: 536 Location: Alberta, Canada.
|
Posted: Tue Mar 18, 2003 10:17 am Post subject: |
|
|
Here is my new attempt to get ntpd running again. My net.eth0 is still running okay throughout this attempt.
Code: |
pine [~]
# nano /etc/conf.d/ntpd
pine [~]
# /etc/init.d/ntpd status
* status: started
pine [~]
# /etc/init.d/ntpd zap
* Manually resetting ntpd to stopped state.
pine [~]
# /etc/init.d/ntpd start
* Initializing clock via ntpdate... [ ok ]
* Starting ntpd...
* Failed to start ntpd [ !! ]
pine [~]
# /etc/init.d/ntpd status
* status: stopped
pine [~]
# /etc/init.d/ntpd zap
pine [~]
# /etc/init.d/ntpd start
* Initializing clock via ntpdate... [ ok ]
* Starting ntpd...
* Failed to start ntpd [ !! ]
pine [~]
# /etc/init.d/ntpd zap
pine [~]
# /etc/init.d/ntpd start
* Initializing clock via ntpdate...
18 Mar 03:03:01 ntpdate[5696]: no servers can be used, exiting
* Failed to run ntpdate [ !! ]
* Starting ntpd...
* Failed to start ntpd [ !! ]
pine [~]
#
|
The first nano was when I added the two other servers to the NTPDATE_OPTS line. That didn't work. So I edited the /etc/conf.d/ntpd file again (in a new non-trackable terminal window) and removed the extra servers leaving only one in the NTPDATE_OPTS line which resulted in the "no servers can be used" error. I can't seem to start ntpd up again.
Here are the latest log entries from /var/log/ntpd.log ...
Code: |
17 Mar 22:39:18 ntpd[5129]: signal_no_reset: signal 17 had flags 4000000
17 Mar 22:43:40 ntpd[5127]: kernel time discipline status change 41
18 Mar 00:27:20 ntpd[5127]: synchronisation lost
18 Mar 00:43:26 ntpd[5127]: time correction of 7191 seconds exceeds sanity limit
(1000); set clock manually to the correct UTC time.
(END)
|
I set the clock manually to the correct UTC time and tried restarting again. No go.
Next I will try to reboot.
I hope you all don't mind me posting a running log of what I am doing but I am hoping someone might see something that I could be doing differently to avoid these problems.
If nothing else works I will take out my AGP video card and replace the clock battery that lies underneath it and see if that helps though I am not at all certain that doing so will help. Since the main purpose of the battery, from what I know is to keep the clock on time while the computer is off.
Thanks.
Carlos _________________ I'm not yet a real Guru so if you are a newbie here and want to teach me a thing or two, please do! I'm still learning just like you. |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
carlos123 Guru
![Guru Guru](/images/ranks/rank_rect_3.gif)
![](images/avatars/gallery/TV Stars/picard1.jpg)
Joined: 12 Feb 2003 Posts: 536 Location: Alberta, Canada.
|
Posted: Tue Mar 18, 2003 10:25 am Post subject: |
|
|
Well, restarting the computer didn't do any good this time. Here is my latest attempt to restart ntpd after the reboot. My network was up and running just fine.
Code: |
pine [~]
# /etc/init.d/ntpd status
* status: started
pine [~]
# /etc/init.d/ntpd stop
* Stopping ntpd... [ ok ]
pine [~]
# /etc/init.d/ntpd zap
pine [~]
# /etc/init.d/ntpd start
* Initializing clock via ntpdate...
18 Mar 10:21:52 ntpdate[1436]: no servers can be used, exiting
* Failed to run ntpdate [ !! ]
* Starting ntpd... [ ok ]
pine [~]
# ntpq -p
remote refid st t when poll reach delay offset jitter
==============================================================================
fsa.cpsc.ucalga subitaneous.cps 2 u 8 64 1 35.555 -251931 0.015
SUE.CC.UREGINA. CLOCK.UREGINA.C 2 u 5 64 1 82.981 -251931 0.015
caesar.cs.wisc. ben.cs.wisc.edu 2 u 13 64 1 79.255 -251931 0.015
|
The problem seems to be ntpdate but am not sure. The boot messages said Gentoo could not start ntpdate during boot up either.
Hmmm....not sure what to try next.
Carlos _________________ I'm not yet a real Guru so if you are a newbie here and want to teach me a thing or two, please do! I'm still learning just like you. |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
bLanark Apprentice
![Apprentice Apprentice](/images/ranks/rank_rect_2.gif)
![](images/avatars/158786021940bc2b56c3d3e.gif)
Joined: 27 Aug 2002 Posts: 181 Location: Royal Berkshire, UK
|
Posted: Tue Mar 18, 2003 10:42 am Post subject: Something else |
|
|
I think that ntpd stops when the local computer time goes too far from the real time. It can't/won't change the time except by a small amount.
So, possibly the real-time clock is very badly broken, or maybe something else is changing the time - are there any other applications running? What does "ps -ef" (or "ps aux") say? Do any of them change the time?
When ntpd fails to start, some servers don't let you connect without authorisation, so it could be that. Or maybe your dns is shot and you can't resolve the names, or maybe your eth0 is broken and so the requests or responses get lost.
With eth0 going down, next time it happens capture the output from dmesg and post it here.
Are you sure you've got the correct kernel module for your eth0? What does it say about your card in /proc/pci ("less /proc/pci")? What does lsmod say, or is support compiled into the kernel?
Sorry, lots of questions, a few ideas but no answers!
bLanark _________________ .sig: access denied |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
carlos123 Guru
![Guru Guru](/images/ranks/rank_rect_3.gif)
![](images/avatars/gallery/TV Stars/picard1.jpg)
Joined: 12 Feb 2003 Posts: 536 Location: Alberta, Canada.
|
Posted: Tue Mar 18, 2003 12:12 pm Post subject: |
|
|
Thanks for your input bLanark. Good questions that you asked.
I don't want to get too heavy into this in the sense of going way deeper than I need to go so I will have to limit myself to working out the answers to those questions that seem to have the best chance of being relevant to the problem.
I have been looking through bunches of NTP documentation on the Internet and it's like going through the U.S. Tax Code. Ridiculously complex!
Just a few comments on your questions. I know the DNS is working because I can ping to the domain names of the servers I am using and name resolution works.
I know too that, although my eth0 service goes down too sometimes, it has been fine during my attempts to get ntpd started again. I can surf just fine and otherwise use the Internet while not being able to start up ntpd.
The output from dmesg is listed on a previous message I posted on this thread.
As for other applicatioins running there are lots of them. I am running inside of KDE and I have KMail, Konqueror, and several Konsoles open while trying to get ntpd started again. Your question does give me an idea though. I will remove kdm as my start up manager and go back to just the command line login and try starting ntpd that way if I can't work it out while using KDE.
I will also have to try and bootup with Redhat and see what options and things are set up there as a reference point. Assuming ntpd is working okay there, if Redhat uses it too. Though it's going to be a bit of a pain to translate the way Redhat does it and apply it to Gentoo's custom init scripting wrapper type of service starting and stopping.
Support for my NIC is not built into the kernel. I load the 3c59x module into the kernel at bootup by placing "3c59x" on a seperate line in the modules.autoload file. If it was not loaded properly I would not be able to surf the Internet.
Your questions have given me some things to track down and try bLanark. I appreciate that.
If you want to know anything else or still believe it would be helpful to see the output from lsmod for example, just let me know and I will post it.
Carlos _________________ I'm not yet a real Guru so if you are a newbie here and want to teach me a thing or two, please do! I'm still learning just like you. |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
carlos123 Guru
![Guru Guru](/images/ranks/rank_rect_3.gif)
![](images/avatars/gallery/TV Stars/picard1.jpg)
Joined: 12 Feb 2003 Posts: 536 Location: Alberta, Canada.
|
Posted: Tue Mar 18, 2003 9:58 pm Post subject: |
|
|
NOTE: I am posting ongoing follow-ups so that those reading this thread and who are new to Gentoo might learn from what I am experiencing. As well as allowing passerby's more opportunity to comment on anything they might see.
Last night, with ntpd down and not running and indeed unable to be run, I logged out of KDE and left the computer with the KDM login manager on the screen.
This morning I logged in as carlos and everything is working just fine. Internet connectivity and ntpd are BOTH up and running!
When looking at the active processes there are three ntpd processes running!! All owned by root.
I am totally and completely without a clue. As to why I could not for the life of me get ntpd started as a su'ed root yesterday and despite a reboot and why today when logging back in, it started on it's own. All I did was log in.
From 4:00 AM last night when I went to bed to now the /var/log/ntpd.log file shows absolutely nothing but ...
Code: |
18 Mar 14:56:10 ntpd[1873]: server returns a permission denied error
|
errors. Nothing else. Yet the clock is updating just fine and ntpd is running. Go figure!
I did find out how to test my battery (for anyone interested in knowing this). Reboot and go into your BIOS. Check the time in the BIOS. Turn off the computer and come back in a few hours (so best done overnight). Go into the BIOS again. If the time is off by more than a minute or two then the battery most probably needs to be replaced.
Carlos _________________ I'm not yet a real Guru so if you are a newbie here and want to teach me a thing or two, please do! I'm still learning just like you. |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
bLanark Apprentice
![Apprentice Apprentice](/images/ranks/rank_rect_2.gif)
![](images/avatars/158786021940bc2b56c3d3e.gif)
Joined: 27 Aug 2002 Posts: 181 Location: Royal Berkshire, UK
|
Posted: Tue Mar 18, 2003 11:11 pm Post subject: |
|
|
Quote: | I have been looking through bunches of NTP documentation on the Internet and it's like going through the U.S. Tax Code. Ridiculously complex! |
Indeed. I'd advise you to forget the detail unless you're writing your own ntp client or daemon.
Quote: | The output from dmesg is listed on a previous message I posted on this thread. |
Yes, but that looks like a dmesg after boot, and before anything goes wrong. I'd hope to see some error message about eth0 later...
Quote: | I know the DNS is working because I can ping to the domain names of the servers I am using and name resolution works. |
OK, the next thing is to see if you can connect to them. Try
Code: |
telnet (servername) 123
|
and see if the connection is refused or not. If it is, then you need to use another server. I use gb.public.ntp.get-time.net - obviously I am in the UK (OK, not so obvious as GB != UK ) - you need to find a local one that will let you connect.
Quote: | Last night, with ntpd down and not running and indeed unable to be run, I logged out of KDE and left the computer with the KDM login manager on the screen.
This morning I logged in as carlos and everything is working just fine. Internet connectivity and ntpd are BOTH up and running! |
This is bizarre. OK, so the "ps" command did NOT list any ntpd when you logged out, but it did when you logged in?
How are you starting ntpd? Is it in your boot scripts - did you do "rc-update add ntpd default", or is it in your .profile or similar? Is it in a daily cron job or something?
The logs (maybe the system log AND /var/log/ntp) MUST show something if the process was not started and now is. Have another look
Perhaps you've been rooted - processes don't start on their own.
bLanark _________________ .sig: access denied |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
carlos123 Guru
![Guru Guru](/images/ranks/rank_rect_3.gif)
![](images/avatars/gallery/TV Stars/picard1.jpg)
Joined: 12 Feb 2003 Posts: 536 Location: Alberta, Canada.
|
Posted: Wed Mar 19, 2003 1:39 am Post subject: |
|
|
Thanks for the additional tips bLanark. Good stuff! I am keeping an eye on it and will be better prepared to see what is happening underneath now that I have a good grasp of what logs to look into, how to check the battery, and so forth. I am assuming that the system log is the /var/log/everything/* logs. If that is not correct I would appreciate a correction of that assumption.
Anyway one way or another I'll post either a fix or a continuing problem once I see something odd again.
As for being rooted - I assume that means some kind of hacking type penetration of my system. It's possible I guess but highly unlikely since I am running a hardware firewall (Smoothwall) that purports to have never had a single instance of penetration between my Linux box and the Internet. My Linux box IP is not even reachable from off the Internet.
Carlos _________________ I'm not yet a real Guru so if you are a newbie here and want to teach me a thing or two, please do! I'm still learning just like you. |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
carlos123 Guru
![Guru Guru](/images/ranks/rank_rect_3.gif)
![](images/avatars/gallery/TV Stars/picard1.jpg)
Joined: 12 Feb 2003 Posts: 536 Location: Alberta, Canada.
|
Posted: Wed Mar 19, 2003 2:39 am Post subject: |
|
|
Well I had to go and mess with ntpd again! Now I can't get it restarted . Just like before. Here is my attempt at documenting my actions. Bear in mind that I started with 3 ntpd related processes running as I had mentioned previously.
Code: |
pine [~] # /etc/init.d/ntpd status
* status: stopped
pine [~] # ps -C ntpd
PID TTY TIME CMD
1861 ? 00:00:00 ntpd
1872 ? 00:00:00 ntpd
1873 ? 00:00:00 ntpd
pine [~] # /etc/init.d/ntpd zap
pine [~] # ps -C ntpd
PID TTY TIME CMD
1861 ? 00:00:00 ntpd
1872 ? 00:00:00 ntpd
1873 ? 00:00:00 ntpd
pine [~] # /etc/init.d/ntpd start
* Starting ntpd...
* Failed to start ntpd [ !! ]
pine [~] # /etc/init.d/ntpd restart
* Starting ntpd...
* Failed to start ntpd [ !! ]
pine [~] # tail /var/log/ntpd.log
18 Mar 19:22:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:22:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:23:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:23:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:24:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:24:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:25:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:25:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:26:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:26:11 ntpd[1873]: server returns a permission denied error
pine [~] # date -u
Wed Mar 19 02:26:42 UTC 2003
pine [~] # date
Tue Mar 18 19:26:46 MST 2003
pine [~] # tail /var/log/everything/current
Mar 18 17:53:19 [su(pam_unix)] session opened for user root by (uid=1000)
Mar 18 17:54:52 [su(pam_unix)] session opened for user root by (uid=1000)
Mar 18 19:15:17 [su(pam_unix)] session opened for user root by (uid=1000)
Mar 18 19:19:09 [rc-scripts] ERROR: "ntpd" has not yet been started.
Mar 18 19:20:35 [rc-scripts] status: stopped
Mar 18 19:20:47 [rc-scripts] ERROR: "ntpd" has not yet been started.
Mar 18 19:23:56 [rc-scripts] status: stopped
Mar 18 19:25:18 [rc-scripts] status: stopped
Mar 18 19:25:51 [rc-scripts] Failed to start ntpd
Mar 18 19:26:09 [rc-scripts] Failed to start ntpd
pine [~] # ps -C ntpd
PID TTY TIME CMD
1861 ? 00:00:00 ntpd
1872 ? 00:00:00 ntpd
1873 ? 00:00:00 ntpd
pine [~] # kill 1861
pine [~] # kill 1872
pine [~] # kill 1873
-bash: kill: (1873) - No such process
pine [~] # ps -C ntpd
PID TTY TIME CMD
pine [~] # tail /var/log/ntpd.log
18 Mar 19:25:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:26:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:26:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:27:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:27:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:28:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:28:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:28:32 ntpd[1861]: ntpd exiting on signal 15
18 Mar 19:28:38 ntpd[1872]: ntpd exiting on signal 15
18 Mar 19:28:41 ntpd[1873]: parent died before we finished, exiting
pine [~] # tail /var/log/everything/current
Mar 18 17:54:52 [su(pam_unix)] session opened for user root by (uid=1000)
Mar 18 19:15:17 [su(pam_unix)] session opened for user root by (uid=1000)
Mar 18 19:19:09 [rc-scripts] ERROR: "ntpd" has not yet been started.
Mar 18 19:20:35 [rc-scripts] status: stopped
Mar 18 19:20:47 [rc-scripts] ERROR: "ntpd" has not yet been started.
Mar 18 19:23:56 [rc-scripts] status: stopped
Mar 18 19:25:18 [rc-scripts] status: stopped
Mar 18 19:25:51 [rc-scripts] Failed to start ntpd
Mar 18 19:26:09 [rc-scripts] Failed to start ntpd
Mar 18 19:27:36 [su(pam_unix)] session opened for user root by (uid=1000)
pine [~] # /etc/init.d/ntpd zap
pine [~] # /etc/init.d/ntpd status
* status: stopped
pine [~] # /etc/init.d/ntpd start
* Starting ntpd...
* Failed to start ntpd [ !! ]
pine [~] # ps -C ntpd
PID TTY TIME CMD
pine [~] # nano /etc/conf.d/ntpd
pine [~] # /etc/init.d/ntpd status
* status: stopped
pine [~] # /etc/init.d/ntpd start
* Initializing clock via ntpdate... [ ok ]
* Starting ntpd...
* Failed to start ntpd [ !! ]
pine [~] # ps -C ntpd
PID TTY TIME CMD
pine [~] # tail /var/log/ntpd.log
18 Mar 19:25:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:26:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:26:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:27:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:27:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:28:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:28:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:28:32 ntpd[1861]: ntpd exiting on signal 15
18 Mar 19:28:38 ntpd[1872]: ntpd exiting on signal 15
18 Mar 19:28:41 ntpd[1873]: parent died before we finished, exiting
pine [~] # tail /var/log/everything/current
Mar 18 19:20:47 [rc-scripts] ERROR: "ntpd" has not yet been started.
Mar 18 19:23:56 [rc-scripts] status: stopped
Mar 18 19:25:18 [rc-scripts] status: stopped
Mar 18 19:25:51 [rc-scripts] Failed to start ntpd
Mar 18 19:26:09 [rc-scripts] Failed to start ntpd
Mar 18 19:27:36 [su(pam_unix)] session opened for user root by (uid=1000)
Mar 18 19:29:38 [rc-scripts] status: stopped
Mar 18 19:29:42 [rc-scripts] Failed to start ntpd
Mar 18 19:31:52 [rc-scripts] status: stopped
Mar 18 19:32:11 [rc-scripts] Failed to start ntpd
pine [~] #
|
The final outcome of all the above? I can't start ntpd again .
One comment on the above. The "nano /etc/conf.d/ntpd" line is when I remembered to uncomment the relevant lines in the /etc/conf.d/ntpd file. Which I did. But as you can see it didn't do any good.
Oh and just prior to the above I also had to turn OFF buffering for the metalog daemon by editing /etc/conf.d/metalog and restarting with "/etc/init.d/metalog restart".
Continuing.....I will try different time servers as you suggested bLanark...
Carlos _________________ I'm not yet a real Guru so if you are a newbie here and want to teach me a thing or two, please do! I'm still learning just like you. |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
carlos123 Guru
![Guru Guru](/images/ranks/rank_rect_3.gif)
![](images/avatars/gallery/TV Stars/picard1.jpg)
Joined: 12 Feb 2003 Posts: 536 Location: Alberta, Canada.
|
Posted: Wed Mar 19, 2003 3:00 am Post subject: |
|
|
I was re-reading through some of the posts in this thread and bLanark something you said about port 123 rung a bell. Here are some log entries from last night from the /var/log/everything/current log file....
Code: |
Mar 18 04:05:07 [ntpd] ntpd 4.1.1b@1.829 Sat Mar 15 16:11:46 MST 2003 (1)
Mar 18 04:05:07 [ntpd] precision = 25 usec
Mar 18 04:05:07 [ntpd] bind() fd 4, family 2, port 123, addr 0.0.0.0, in_classd=
0 flags=1 fails: Address already in use
Mar 18 04:05:07 [ntpd] bind() fd 4, family 2, port 123, addr 127.0.0.1, in_class
d=0 flags=0 fails: Address already in use
Mar 18 04:05:07 [ntpd] bind() fd 4, family 2, port 123, addr 192.168.1.100, in_c
lassd=0 flags=1 fails: Address already in use
Mar 18 04:05:07 [ntpd] kernel time discipline status 0040
Mar 18 05:05:12 [su(pam_unix)] session opened for user root by (uid=1000)
Mar 18 05:22:53 [su(pam_unix)] session closed for user root
|
Not sure what they mean but they seem somewhat problematic and they have to do with port 123. Anybody want to take a shot and translate it into English for me ?
Carlos _________________ I'm not yet a real Guru so if you are a newbie here and want to teach me a thing or two, please do! I'm still learning just like you. |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
carlos123 Guru
![Guru Guru](/images/ranks/rank_rect_3.gif)
![](images/avatars/gallery/TV Stars/picard1.jpg)
Joined: 12 Feb 2003 Posts: 536 Location: Alberta, Canada.
|
Posted: Wed Mar 19, 2003 9:53 am Post subject: |
|
|
Well wouldn't you know it? ntpd works just fine under Redhat!
Here's are some actions on ntpd running under Redhat:
Code: |
[root@localhost log]# /etc/init.d/ntpd stop
ntpd: Removing firewall opening for port 123 [ OK ]
Shutting down ntpd: [ OK ]
[root@localhost log]# /etc/init.d/ntpd start
ntpd: Opening firewall for port 123 [ OK ]
ntpd: Synchronizing with time server: [ OK ]
Starting ntpd: [ OK ]
[root@localhost log]# ps -C ntpd
PID TTY TIME CMD
1201 ? 00:00:00 ntpd
1204 ? 00:00:00 ntpd
[root@localhost log]# /etc/init.d/ntpd stop
ntpd: Removing firewall opening for port 123 [ OK ]
Shutting down ntpd: [ OK ]
[root@localhost log]# ps -C ntpd
PID TTY TIME CMD
[root@localhost log]# /etc/init.d/ntpd start
ntpd: Opening firewall for port 123 [ OK ]
ntpd: Synchronizing with time server: [ OK ]
Starting ntpd: [ OK ]
[root@localhost log]#
|
The main difference between the Redhat /etc/init.d/ntpd code (yes Redhat has this file too though their code is way different) is that their code does things to open and close port 123 using iptables.
Now the problem is that whatever they do or don't do with firewall iptables it does NOT affect my hardware firewall running on a totally different computer. So if they are succeeding at what Gentoo is not the problem comes right back to something on my Gentoo system.
In other words I do not believe that I need to change anything on the -real- firewall, my hardware firewall computer, if Redhat is able to use ntpd just fine. Something on my Gentoo system is just not right!
I'll keep plugging away at this I guess.
You know for all it's faults with RPM's and otherwise Redhat sure does know how to make things easy on the user. Such that one can start getting some productive work out of their computer without hundreds of hours of tinkering. Alas...if only they had portage and this fine forum!
Carlos _________________ I'm not yet a real Guru so if you are a newbie here and want to teach me a thing or two, please do! I'm still learning just like you. |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
carlos123 Guru
![Guru Guru](/images/ranks/rank_rect_3.gif)
![](images/avatars/gallery/TV Stars/picard1.jpg)
Joined: 12 Feb 2003 Posts: 536 Location: Alberta, Canada.
|
Posted: Wed Mar 19, 2003 9:32 pm Post subject: |
|
|
This will be my last post on this thread since I seem to be only talking to myself. The thread seems to have sunk into obscurity.
Last night I did not log out of KDE and left the computer running just fine. With both net.eth0 and ntpd services running.
This morning I come back and both are out again!! No apparent reason. Nothing! Fortunately I was able to zap the ntpd service and restart it again. Here is a recording of my attempt.
Code: |
pine [~] # tail /var/log/ntpd.log
18 Mar 19:27:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:27:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:28:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:28:11 ntpd[1873]: server returns a permission denied error
18 Mar 19:28:32 ntpd[1861]: ntpd exiting on signal 15
18 Mar 19:28:38 ntpd[1872]: ntpd exiting on signal 15
18 Mar 19:28:41 ntpd[1873]: parent died before we finished, exiting
19 Mar 04:01:19 ntpd[1021]: frequency initialized 1.926 from /var/lib/misc/ntp.drift
19 Mar 04:01:19 ntpd[1056]: signal_no_reset: signal 17 had flags 4000000
19 Mar 04:02:05 ntpd[1021]: kernel time discipline status change 41
pine [~] # tail /var/log/everything/current
Mar 19 04:02:49 [kde(pam_unix)] session closed for user carlos
Mar 19 04:02:50 [pam_console] getgrnam failed for cdwriter
Mar 19 04:03:03 [kde(pam_unix)] session opened for user carlos by (uid=0)
Mar 19 04:07:49 [su(pam_unix)] session opened for user root by (uid=1000)
Mar 19 06:10:14 [su(pam_unix)] session opened for user root by (uid=1000)
Mar 19 06:10:14 [su(pam_unix)] session closed for user root
Mar 19 06:10:14 [su(pam_unix)] session opened for user root by (uid=1000)
Mar 19 06:10:46 [su(pam_unix)] session closed for user root
Mar 19 06:36:30 [su(pam_unix)] session closed for user root
Mar 19 07:48:17 [su(pam_unix)] session opened for user root by (uid=1000)
pine [~] # ps -C ntpd
PID TTY TIME CMD
1021 ? 00:00:00 ntpd
pine [~] # /etc/init.d/ntpd status
* status: started
pine [~] # ntpq -p
remote refid st t when poll reach delay offset jitter
==============================================================================
pine [~] # /etc/init.d/net.eth0 status
* status: started
pine [~] # ping www.google.com
PING www.google.com (216.239.57.100): 56 octets data
--- www.google.com ping statistics ---
5 packets transmitted, 0 packets received, 100% packet loss
pine [~] # /etc/init.d/net.eth0 stop
* Unmounting network filesystems... [ ok ]
* Stopping ntpd... [ ok ]
* Bringing eth0 down... [ ok ]
pine [~] # /etc/init.d/net.eth0 start
* Bringing eth0 up... [ ok ]
* Setting default gateway... [ ok ]
pine [~] # ping www.google.com
PING www.google.com (216.239.57.100): 56 octets data
64 octets from 216.239.57.100: icmp_seq=0 ttl=50 time=99.5 ms
64 octets from 216.239.57.100: icmp_seq=1 ttl=50 time=98.4 ms
64 octets from 216.239.57.100: icmp_seq=2 ttl=50 time=97.6 ms
64 octets from 216.239.57.100: icmp_seq=3 ttl=50 time=97.5 ms
--- www.google.com ping statistics ---
4 packets transmitted, 4 packets received, 0% packet loss
round-trip min/avg/max = 97.5/98.2/99.5 ms
pine [~] # ntpq -p
ntpq: read: Connection refused
pine [~] # /etc/init.d/ntpd status
* status: stopped
pine [~] # /etc/init.d/ntpd stop
* ERROR: "ntpd" has not yet been started.
pine [~] # /etc/init.d/ntpd zap
pine [~] # ps -C ntpd
PID TTY TIME CMD
pine [~] # /etc/init.d/ntpd start
* Initializing clock via ntpdate... [ ok ]
* Starting ntpd... [ ok ]
pine [~] # ps -C ntpd
PID TTY TIME CMD
3190 ? 00:00:00 ntpd
pine [~] # date
Wed Mar 19 14:27:26 MST 2003
pine [~] # date -u
Wed Mar 19 21:27:33 UTC 2003
pine [~] # ntpq -p
remote refid st t when poll reach delay offset jitter
==============================================================================
fsa.cpsc.ucalga montpelier.ilan 2 u 15 64 1 827.097 -48.658 0.015
SUE.CC.UREGINA. CLOCK.UREGINA.C 2 u 2 64 1 86.039 -435.30 4.982
pine [~] #
|
What is odd in the log output shown up is this....
Code: |
Mar 19 06:36:30 [su(pam_unix)] session closed for user root
Mar 19 07:48:17 [su(pam_unix)] session opened for user root
|
I exited a root terminal window last night at 6:36 AM and when I came back this "morning" about 8 hours later the clock said 7:48 (I am assuming AM).
If anyone has any idea as to what might be causing Gentoo to act so strangely I would appreciate hearing it. This problem appears to be Gentoo related since no such problems crop up when booting Redhat on the same computer!
Thanks.
Carlos _________________ I'm not yet a real Guru so if you are a newbie here and want to teach me a thing or two, please do! I'm still learning just like you. |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
bLanark Apprentice
![Apprentice Apprentice](/images/ranks/rank_rect_2.gif)
![](images/avatars/158786021940bc2b56c3d3e.gif)
Joined: 27 Aug 2002 Posts: 181 Location: Royal Berkshire, UK
|
Posted: Wed Mar 19, 2003 11:31 pm Post subject: |
|
|
workaround: Why not just set up an hourly cron job to set the date via ntpdate, and forget about the daemon?
Why do you need ntpd anyway? Is your clock inaccurate?
I still suspect that something else is changing the time on your machine as the "kernel time discipline status change 41" error suggests that the time is too far out to change. If this always happens at a particular time, it is probably a running daemon or a cron job.
What happens (to ntpd) if you log out of the box entirely (no display manager either) for a night? What happens if you leave a normal user or root logged in on a console?
(That'll take you 3 days !)
BTW: the "time correction of 7191 seconds exceeds sanity limit" is approx 2 hours. That's a bug drift .... is your timezone set correctly everywhere?
bLanark _________________ .sig: access denied |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
carlos123 Guru
![Guru Guru](/images/ranks/rank_rect_3.gif)
![](images/avatars/gallery/TV Stars/picard1.jpg)
Joined: 12 Feb 2003 Posts: 536 Location: Alberta, Canada.
|
Posted: Thu Mar 20, 2003 3:17 am Post subject: |
|
|
Well I guess it's not my last post on this thread after all .
Thanks again bLanark!! You've given me some more good input to work my way through.
One reason that I want to use ntpd is that I HATE leaving loose ends and having to tell myself that I cannot otherwise use a piece of perfectly good software on Gentoo when I can use it on other systems like Redhat. To me that spells defeat for Gentoo in my mind and undoubtedly in the minds of others that I recommend Gentoo to.
I mean how would it sound for me to say something like "Yeah Gentoo is a great system and much better than Redhat but...let's see...you can't run ntpd like you can under Redhat"
So my customer asks me why not to which I fiddle and faddle and look at my shoes and say "Well...I really don't know. You just can't". Not only is that an embarrassment to Gentoo but to me as being too incompetent to figure it out. At least in the mind of my customers. Regardless of the reality.
I know that it might seem a bit picky but in encouraging others to try out Linux (and Gentoo) every little such pickiness can make or break the willingness of others to try it. And to pay me to install it on their systems. Or to support them in it's use.
I want to shield the average customer from all the gibberish and all the problems and all the configuration headaches that I have experienced since coming to Linux. Frankly for all the good press I heard about Linux I still can't believe all the hassles associated with it's use (it took me something like two days just to get my USB mouse to work under Gentoo!). My customers just want to be productive and to receive a working system. Not to become Linux hackers. And I so much want to be a part of getting people away from Microsoft and to experience freedom through Linux.
Not because I hate Microsoft but because Linux is trully a great OS once you work your way through all the gibberish, configuration, and software installation hassles that hold it back from mainstream acceptance. At least on the Desktop.
Another reason bLanark is that I don't want to concern myself with how accurate the clock is being. I have been reading a massive thread on this forum about clocks loosing or gaining time under Gentoo (95 posts last I counted). Using ntpd will prevent this for me. Where I won't even have to pay attention to the clock and be left wondering if the clock is accurate today as oppossed to yesterday type of thing. If ntpd is running then it will be accurate.
I want a dependable clock. I *need* a dependable clock. So do my customers. I can't afford to have a Gentoo system clock to be accurate one day and loose or gain time the next.
The thread in question is a real eye opener for me in terms of not depending on the clock alone.
In terms of just sticking ntpdate in a cron job...ntpdate is a deprecated utility which the ntpd web site recommends one's not using anymore. The functionality of ntpdate has been moved entirely into ntpd. It may work...for now. But it's deprecated.
And from my CSS, HTML, and Perl background if something is deprecated I no longer want to use it. If I can help it. Just a pet peeve of mine I guess but then again that's why these projects label things as deprecated. So that users will no longer use the deprecated item.
Which brings me right back to why ntpd won't run consistently and accurately only under Gentoo.....I just don't know yet but I'm not going to give up on this.
Aside from what you have suggested I will be adding my own output comments to Gentoo source code to better tell where the attempt to run ntpd is at (when it refuses to load). I guess you could say that I am going to do a little hacking and debugging of the source code .
Anyway I'll post my results back to this thread I guess when I have implemented some of your suggestions and tried some other things.
Thanks again bLanark for helping me out with your input. Very appreciated. I hope I didn't come accross as too frustrated. It wasn't personally directed at you.
It's just that lately Linux and more specifically Gentoo has been such an incredible pain in the butt to deal with! It's like something always needs some tweaking or configuration or patching.
I'm just looking forward to the day when I have a solid and working Gentoo system that doesn't need so much of this. I am truly hoping that there is such a day ahead and that I won't come to find out that using Linux is a continual wading through gibberish and just more hacking and tinkering and attempts to keep the whole thing running smoothly.
I keep talking Linux up as being so much better than Windows but a growing and gnawing doubt is beginning to surface within me that maybe it isn't. Windows, for all it's flaws, is so much more imminently useful right out of the box in comparision to Linux (or at least Gentoo) that sometimes I wonder if I made the right choice switching to Linux.
For me, I've spent too much time in Linux to throw it all away and go back. But for my customers? I'm not so sure that I will recommend Linux as much as I would have before I switched myself and came to know it hands on. It's a great OS for certain uses but overall it's one incredible pain to work with!
Thanks for "listening". I needed to get that off my chest.
Carlos _________________ I'm not yet a real Guru so if you are a newbie here and want to teach me a thing or two, please do! I'm still learning just like you. |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
carlos123 Guru
![Guru Guru](/images/ranks/rank_rect_3.gif)
![](images/avatars/gallery/TV Stars/picard1.jpg)
Joined: 12 Feb 2003 Posts: 536 Location: Alberta, Canada.
|
Posted: Wed Mar 26, 2003 11:37 pm Post subject: |
|
|
If you are still "listening" in to this thread bLanark, just wanted to let you know that I tried variations of both things that you suggested. The problem was still there when I changed the display manager to xdm instead of kdm and logged in to a console login prompt as root (although I did not try just keeping the computer logged out).
A BIOS check of the time was fine overnight so it wasn't the battery.
The odd thing is that the problem has apparently disappeared. Entirely!! My net connection and my ntpd service stays on always now. Like it should. And I have not changed the time servers queried at all.
It's as if the problem was never there.
I am just going to enjoy the working system for now and not bother to fiddle with this problem again unless it pops up again. I did emerge some other programs and do other things on my system and it is possible that something that was screwy got fixed in the process. But as to what that is? Who knows.
I suspect that I will eventually face this problem again. If not on my own computer then on that of my customers computers that I install Gentoo on. But I guess I will cross that bridge again when I come to it
Thanks again for your input and help bLanark. Much appreciated.
Carlos |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
bLanark Apprentice
![Apprentice Apprentice](/images/ranks/rank_rect_2.gif)
![](images/avatars/158786021940bc2b56c3d3e.gif)
Joined: 27 Aug 2002 Posts: 181 Location: Royal Berkshire, UK
|
Posted: Thu Mar 27, 2003 7:30 am Post subject: |
|
|
I am still listening. Thanks for the update - very strange, but if it ain't broken, don't fix it!
I'm a little wary about installing gentoo at customer sites - I'd rather rely on redhat (apart from the limited lifetime of each version) or similar, and have something that is tested a little more than what's in portage. I've disabled my machine on countless occasions just by "emerge rsync; emerge -u --deep world".
For a *big* shop, however, I can see gentoo being good, where there are dozens of identical machines and they have the resources to test everything thoroughly before putting it on the desktops (or whatever). It'd be nice to have a customer that big, however.
bLanark _________________ .sig: access denied |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
carlos123 Guru
![Guru Guru](/images/ranks/rank_rect_3.gif)
![](images/avatars/gallery/TV Stars/picard1.jpg)
Joined: 12 Feb 2003 Posts: 536 Location: Alberta, Canada.
|
Posted: Thu Mar 27, 2003 1:34 pm Post subject: |
|
|
Yeah I am still leery myself about putting Gentoo on customers machines myself. Precisely because of these types of quirks that I have experienced.
If things work out I will shortly be giving free seminars on why one might want to go with Linux (in a joint venture with my computer supplier) but have not yet decided exactly how I am going to supply those who want Linux with a working distro. Aside from charging them for an extra hard disk, a whole system, or for the labor to install it of course .
Redhat is SO.o.o.o.o much easier to install and initially configure than Gentoo. And it has fantastic hardware detection. Yet Gentoo has portage which makes installation of programs and the latest versions SO.o.o.o.o much easier than on Redhat (albeit still a bit problematic with it's own quirks).
I wish there was a Redhatted type of Gentoo with portage . I may just compile programs under Gentoo and copy them over or make them available to those of my customers who want the binaries. Assuming that is even possible. While installing Redhat on their computers. I am not sure yet.
I might follow the lead of Lindows Linux here and offer to provide binaries to customers for a small fee per month which will include my doing the edit the configuration file thingy over VNC or some such program.
My goal will be to capitalize on the growing curiosity about Linux by offering my services to overcome the negatives associated with Linux. Mainly the installation of it and subsequent package installation and configuration.
The quirks of Gentoo of which there are several are definitely frustrating and very time consuming to deal with. Redhat just works! And works reasonably well. Right out of the box. Without quirks (that I have experienced at least).
Carlos |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|
|