Monitorix
Monitorix copied to clipboard
mikaku
Use of uninitialized value in multiplication in int.pm
I just updated Monitorix and I have a new message in the logs :
Use of uninitialized value in multiplication (*) at /usr/lib/monitorix/int.pm line 1175.
Use of uninitialized value in multiplication (*) at /usr/lib/monitorix/int.pm line 1176.
1174 if($version eq "new") {
1175 $picz_width = $picz->{image_width} * $config->{global_zoom};
1176 $picz_height = $picz->{image_height} * $config->{global_zoom};
1177 } else {
1178 $picz_width = $width + 115;
1179 $picz_height = $height + 100;
1180 }
If it can help, since the beginning the graph int1.1day.png works but the image for in2.1day.png is not generated without errors messages (not found).
Please, make sure you have defined the option global_zoom in your configuration file. If not, then it means its time to update that configuration file.
The version of RRDtool is RRDtool 1.4.7 and I have this options defined:
enable_zoom = y
global_zoom = 1.2
OK, thanks, I'm using 1.3.8 here, I've not tested it in 1.4 branch.
Those error messages Use of uninitialized value in multiplication (*) at /usr/lib/monitorix/int.pm line 1175. also appear in the rest of graphs or only in this one (the Interrupts graph)?
I have only this error message after updated all files, and since the beginning the graph int1 is working but int2 not appear (without error log message). There is a way to test only int.pm from the command line ?
I've just download the ZIP file of the master code at Github and tested it on a CentOS 7 x86_64 with RRDtool v1.4.7 and it works finely and without any error message.
Make sure int2 already not appeared before you upgraded to the current master code at Github. Interrupts graph is an old code (still pending to a complete rewrite) that might not easily understand all architectures and it could fail when trying to show all device interrupts.
There is a way to test only int.pm from the command line ?
No, that's not possible.
Please, make sure you installed all the Monitorix components correctly in your system.
I've noticed this isssue on a fresh Debain 7 x32 install using the apt repo from izzysoft. These errors are only seen in the log file. Graph appears to render correctly. RRDtool 1.4.7, Monitorix version 3.7.0 (12-Mar-2015).
/etc/monitorix/monitorix.conf
enable_zoom = y
global_zoom = 1
/var/log/monitorix
HTTPServer: You can connect to your server at http://localhost:8080/
Use of uninitialized value in multiplication (*) at /usr/lib/monitorix/int.pm line 1175.
Use of uninitialized value in multiplication (*) at /usr/lib/monitorix/int.pm line 1176.
Hi PMaynard,
I think that by any reason your second interrupt graph is not rendering, returning uninitialized values from the RRDtool function and ending up showing error messages when trying to multiply them for any value.
As I told in one of my previous posts, the interrupt graph has old code (still pending to a complete rewrite) that might not easily understand all architectures and it could fail when trying to show all device interrupts.
If you don't mind, you may paste the output of cat /proc/interrupts and I'll see what can I do.
Thanks.
Hi Jordi,
If it can help for future improvments, this is my output of cat /proc/interrupts. The system is Ubuntu Trusty on a XenServer virtual machine. Thanks.
CPU0 CPU1 CPU2 CPU3
16: 837307248 0 0 0 xen-percpu-virq timer0
17: 1744 0 0 0 xen-percpu-ipi spinlock0
18: 9330690 0 0 0 xen-percpu-ipi resched0
19: 2940 0 0 0 xen-percpu-ipi callfunc0
20: 0 0 0 0 xen-percpu-virq debug0
21: 588030 0 0 0 xen-percpu-ipi callfuncsingle0
22: 1336152 0 0 0 xen-percpu-ipi irqwork0
23: 0 221059732 0 0 xen-percpu-virq timer1
24: 0 9715 0 0 xen-percpu-ipi spinlock1
25: 0 13404923 0 0 xen-percpu-ipi resched1
26: 0 2859 0 0 xen-percpu-ipi callfunc1
27: 0 0 0 0 xen-percpu-virq debug1
28: 0 564311 0 0 xen-percpu-ipi callfuncsingle1
29: 0 838455 0 0 xen-percpu-ipi irqwork1
30: 0 0 285493562 0 xen-percpu-virq timer2
31: 0 0 9782 0 xen-percpu-ipi spinlock2
32: 0 0 11308637 0 xen-percpu-ipi resched2
33: 0 0 2742 0 xen-percpu-ipi callfunc2
34: 0 0 0 0 xen-percpu-virq debug2
35: 0 0 638590 0 xen-percpu-ipi callfuncsingle2
36: 0 0 395541 0 xen-percpu-ipi irqwork2
37: 0 0 0 259746856 xen-percpu-virq timer3
38: 0 0 0 9241 xen-percpu-ipi spinlock3
39: 0 0 0 11054251 xen-percpu-ipi resched3
40: 0 0 0 2295 xen-percpu-ipi callfunc3
41: 0 0 0 0 xen-percpu-virq debug3
42: 0 0 0 550374 xen-percpu-ipi callfuncsingle3
43: 0 0 0 360597 xen-percpu-ipi irqwork3
44: 424128 0 0 0 xen-dyn-event xenbus
45: 49 0 0 0 xen-dyn-event hvc_console
46: 3087279 0 0 0 xen-dyn-event blkif
47: 5510585 0 0 0 xen-dyn-event eth0
NMI: 0 0 0 0 Non-maskable interrupts
LOC: 0 0 0 0 Local timer interrupts
SPU: 0 0 0 0 Spurious interrupts
PMI: 0 0 0 0 Performance monitoring interrupts
IWI: 1336152 838455 395541 360597 IRQ work interrupts
RTR: 0 0 0 0 APIC ICR read retries
RES: 9330690 13404923 11308637 11054251 Rescheduling interrupts
CAL: 590970 567170 641332 552669 Function call interrupts
TLB: 0 0 0 0 TLB shootdowns
TRM: 0 0 0 0 Thermal event interrupts
THR: 0 0 0 0 Threshold APIC interrupts
MCE: 0 0 0 0 Machine check exceptions
MCP: 0 0 0 0 Machine check polls
ERR: 0
MIS: 0
Dominique,
Thanks for sharing this information, it indeed will be useful when rewriting the new Interrupt graph which I plan to start it soon.
I'll keep you informed of my progressions here in. Thanks again.
@mikaku You are correct I was missing a second interrupt graph, didn't see it fail. Virtual machine running on a xen server, Debain 7 x32.
/proc/interrupts
CPU0 CPU1 CPU2
16: 20540 0 0 xen-percpu-virq timer0
17: 8665 0 0 xen-percpu-ipi resched0
18: 237 0 0 xen-percpu-ipi callfunc0
19: 0 0 0 xen-percpu-virq debug0
20: 624 0 0 xen-percpu-ipi callfuncsingle0
21: 0 18212 0 xen-percpu-virq timer1
22: 0 12042 0 xen-percpu-ipi resched1
23: 0 396 0 xen-percpu-ipi callfunc1
24: 0 0 0 xen-percpu-virq debug1
25: 0 817 0 xen-percpu-ipi callfuncsingle1
26: 0 0 22310 xen-percpu-virq timer2
27: 0 0 9622 xen-percpu-ipi resched2
28: 0 0 433 xen-percpu-ipi callfunc2
29: 0 0 0 xen-percpu-virq debug2
30: 0 0 855 xen-percpu-ipi callfuncsingle2
31: 9863 0 0 xen-dyn-event vfb
33: 0 0 0 xen-dyn-event vkbd
34: 17110 0 0 xen-dyn-event blkif
35: 168 0 0 xen-dyn-event blkif
36: 8350 0 0 xen-dyn-event eth0
176: 800 0 0 xen-dyn-event xenbus
NMI: 0 0 0 Non-maskable interrupts
LOC: 0 0 0 Local timer interrupts
SPU: 0 0 0 Spurious interrupts
PMI: 0 0 0 Performance monitoring interrupts
IWI: 0 0 0 IRQ work interrupts
RES: 8665 12042 9622 Rescheduling interrupts
CAL: 861 1213 1288 Function call interrupts
TLB: 0 0 0 TLB shootdowns
TRM: 0 0 0 Thermal event interrupts
THR: 0 0 0 Threshold APIC interrupts
MCE: 0 0 0 Machine check exceptions
MCP: 0 0 0 Machine check polls
ERR: 0
MIS: 0
P.S Great work with Monitorix.
PMaynard,
Thanks for your contribution. It looks pretty similar to Dominique output so that's probably why you both are experiencing similar problems. It will be indeed useful when writing the new Interrupts graph.
Thanks again and glad to know you are enjoying Monitorix.
@mikaku
hi, if got the same error on actuall archlinux. Installed monitorix 3.7.0 with yaourt monitorix
1 aur/monitorix 3.7.0-1 [installed] (74)
A lightweight system monitoring tool that uses rrd databases.
My output of: cat /proc/interrupts cat /proc/interrupts
CPU0 CPU1
0: 22 0 IO-APIC-edge timer
1: 3 0 IO-APIC-edge i8042
8: 1 0 IO-APIC-edge rtc0
9: 3 0 IO-APIC-fasteoi acpi
12: 4 0 IO-APIC-edge i8042
16: 49 23 IO-APIC 16-fasteoi ehci_hcd:usb1
23: 37 32 IO-APIC 23-fasteoi ehci_hcd:usb4
24: 0 0 PCI-MSI-edge xhci_hcd
25: 0 0 PCI-MSI-edge xhci_hcd
26: 0 0 PCI-MSI-edge xhci_hcd
27: 36174 6511 PCI-MSI-edge 0000:00:1f.2
28: 15 0 PCI-MSI-edge mei_me
29: 147856 20376 PCI-MSI-edge extern0
30: 9 0 PCI-MSI-edge intern0
31: 990 2 PCI-MSI-edge snd_hda_intel
32: 1488 903 PCI-MSI-edge i915
NMI: 43 43 Non-maskable interrupts
LOC: 820836 805319 Local timer interrupts
SPU: 0 0 Spurious interrupts
PMI: 43 43 Performance monitoring interrupts
IWI: 0 0 IRQ work interrupts
RTR: 0 0 APIC ICR read retries
RES: 23137 34211 Rescheduling interrupts
CAL: 3014 458 Function call interrupts
TLB: 9162 8427 TLB shootdowns
TRM: 0 0 Thermal event interrupts
THR: 0 0 Threshold APIC interrupts
MCE: 0 0 Machine check exceptions
MCP: 29 29 Machine check polls
HYP: 0 0 Hypervisor callback interrupts
ERR: 0
MIS: 0
No graph pictures are generated but a lot of apache error logs messages:
[Wed Apr 22 20:58:45.822692 2015] [cgi:error] [pid 5675] [client 10.3.x.x:54329] AH01215: Use of uninitialized value in multiplication (*) at /usr/lib/monitorix/user.pm line 425.: /srv/http/monitorix/cgi/monitorix.cgi
@mugnox,
Thank you for sharing this information, this indeed puts more pressure on me to fix all these issues in the Interrupts graph :)
Regards.
@mikaku Hi Jordi,I encounter this error too it seems.
Monitorix 1.7 on Ubuntu mate 15.04 for Raspberry pi 2 rrdtool 1.4.8 /etc/monitorix/monitorix.conf enable_zoom = y global_zoom = 1
tail /varlog/monitorix
Argument "IPI0" isn't numeric in numeric lt (<) at /usr/lib/monitorix/int.pm line 1015.
Argument "IPI0" isn't numeric in numeric gt (>) at /usr/lib/monitorix/int.pm line 873, <IN> line 15.
Argument "IPI1" isn't numeric in numeric gt (>) at /usr/lib/monitorix/int.pm line 873, <IN> line 16.
Argument "IPI2" isn't numeric in numeric gt (>) at /usr/lib/monitorix/int.pm line 873, <IN> line 17.
Argument "IPI3" isn't numeric in numeric gt (>) at /usr/lib/monitorix/int.pm line 873, <IN> line 18.
Argument "IPI4" isn't numeric in numeric gt (>) at /usr/lib/monitorix/int.pm line 873, <IN> line 19.
Argument "IPI5" isn't numeric in numeric gt (>) at /usr/lib/monitorix/int.pm line 873, <IN> line 20.
Argument "IPI6" isn't numeric in numeric gt (>) at /usr/lib/monitorix/int.pm line 873, <IN> line 21.
Argument "IPI7" isn't numeric in numeric gt (>) at /usr/lib/monitorix/int.pm line 873, <IN> line 22.
Argument "IPI0" isn't numeric in numeric lt (<) at /usr/lib/monitorix/int.pm line 1015.
cat /proc/interrupts
CPU0 CPU1 CPU2 CPU3
16: 0 0 0 0 ARMCTRL bcm2708_fb dma
24: 0 0 0 0 ARMCTRL DMA IRQ
25: 717 0 0 0 ARMCTRL DMA IRQ
32: 753299 0 0 0 ARMCTRL dwc_otg, dwc_otg_pcd, dwc_otg_hcd:usb1
52: 0 0 0 0 ARMCTRL BCM2708 GPIO catchall handler
65: 536 0 0 0 ARMCTRL ARM Mailbox IRQ
66: 4371 0 0 0 ARMCTRL VCHIQ doorbell
75: 1 0 0 0 ARMCTRL
79: 0 0 0 0 ARMCTRL bcm2708_i2c.1
80: 0 0 0 0 ARMCTRL bcm2708_spi.0
84: 3551 0 0 0 ARMCTRL mmc0
99: 99040 74462 65476 52507 ARMCTRL arch_timer
FIQ: usb_fiq
IPI0: 0 0 0 0 CPU wakeup interrupts
IPI1: 0 0 0 0 Timer broadcast interrupts
IPI2: 47229 55686 51337 41427 Rescheduling interrupts
IPI3: 8 73 73 74 Function call interrupts
IPI4: 1922 3089 2411 2642 Single function call interrupts
IPI5: 0 0 0 0 CPU stop interrupts
IPI6: 16 0 0 0 IRQ work interrupts
IPI7: 0 0 0 0 completion interrupts
Err: 0
I hope this helps. Thanks for all your efforts. Bénoît
@BenoitSvB
Thanks for your feedback, I'll put hands to work as soon as possible. Regards.
@mikaku, Thanks for making such a great monitoring tool. I am facing the same issue on my Raspberry Pi, running Ubuntu Mate 16.04, with kernel 4.4.14-v7+. Because of the huge amount of logging, I guess that this will reduce the life of my SD.... Could you please give it a higher prio, please? Regards, Jacco
@Haringstad,
Yeah, I know I'm a bit disconnected but I'm pretty busy at work lately.
I have in mind to change the way how Monitorix shows the interrupt graph, and instead of trying to detect architectures automatically (which resulted pretty difficult) I'd like to include a new option like arch=.
Possible values could be: pc-standard, xen, rpi, or even rpv3, etc. This way, Monitorix can have a predefined set of different architectures and the graph layout should have better results.
I'm doing my best to have enough time for Monitorix, believe me. Thanks for your patience.
You are most welcome! For now, I only let it run when I suspect that something is going on. So, for now, I do have a working solution!
You make sure that you take care of yourself!
And, it seems like a great plan you have, to use the arch= option!
Regards,
Jacco van Koll
On Tue, Jul 26, 2016 at 1:12 PM, Jordi Sanfeliu [email protected] wrote:
@Haringstad https://github.com/Haringstad,
Yeah, I know I'm a bit disconnected but I'm pretty busy at work lately.
I have in mind to change the way how Monitorix shows the interrupt graph, and instead of trying to detect architectures automatically (which resulted pretty difficult) I'd like to include a new option like arch=.
Possible values could be: pc-standard, xen, rpi, or even rpv3, etc. This way, Monitorix can have a predefined set of different architectures and the graph layout should have better results.
I'm doing my best to have enough time for Monitorix, believe me. Thanks for your patience.
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/mikaku/Monitorix/issues/77#issuecomment-235238453, or mute the thread https://github.com/notifications/unsubscribe-auth/AHeCZo_iDDB0F4qK01RgwukKf8UEOFibks5qZeuxgaJpZM4Dhxfx .