[Nut-upsuser] Segmentation fault
Hi! I'm running NUT 2.0.4 and occasionally I get a segfault when starting newhidups and if I don't get a segfault when starting newhidups the process starts eating away at my memory. I sat down trying to genererate -D debug info, but just because of that I can't get it to do a segfault. So until I can, has things changed in the development version, that might have fixed this? Network UPS Tools: New USB/HID UPS driver 0.28 (2.0.4) Detected a UPS: American Power Conversion/Back-UPS CS 500 FW:808.q7.I USB FW:q7 Using subdriver: APC/CyberPower HID 0.9 ___ Nut-upsuser mailing list Nut-upsuser@lists.alioth.debian.org http://lists.alioth.debian.org/mailman/listinfo/nut-upsuser
Re: [Nut-upsuser] Sweex 1000VA UPS
On 11/10/06, Arjen de Korte [EMAIL PROTECTED] wrote:[...] If you're willing to try a few more things, comment out the part in theNUT startup script where 'upsmon' is started/stopped and fire it up. With'upsc [EMAIL PROTECTED]' you should be able to see the status of the UPS, without the risk of shutting down your system when things are notconfigured properly. Try at least /dev/ttyS[0-3] with the UPS in differentserial ports on your server. Remember to stop the script before modifying and starting it again. Good luck!Kind regards,ArjenOk, I 've done some more testing. I did comment out the upsmon entry in /etc/default/nutThen loaded the driver like: /lib/nut/genericups -x upstype=7 /dev/ttyS0It loaded and according syslog:Nov 11 10:19:47 radius1 genericups[8469]: Startup successfulHowever when I tried: radius1:/etc/nut# upsc [EMAIL PROTECTED] Error: Connection failure: Connection refusedradius1:/etc/nut# I've done this for ttyS0 ttyS1 (there are no more, 2 results in an error at fire up)I also changed the permissions on ttyS1 like radius1:/etc/nut# ls -al /dev/ttyS0crw-rw 1 root nut 4, 64 2005-02-26 07:39 /dev/ttyS0radius1:/etc/nut# Same problem.The only things left here are.1. There is a problem with the serial cable. (I tried 2 of them) How will I know if the cable that came with the UPS is the right one?2. That thing just won't ever work for linux.The new series of MGE are Ellipse ASR I hope someone can confirm these things work as they should with NUT running on debian stable.Many tnxYves ___ Nut-upsuser mailing list Nut-upsuser@lists.alioth.debian.org http://lists.alioth.debian.org/mailman/listinfo/nut-upsuser
Re: [Nut-upsuser] Sweex 1000VA UPS
Ok, I 've done some more testing. I did comment out the upsmon entry in /etc/default/nut Then loaded the driver like: /lib/nut/genericups -x upstype=7 /dev/ttyS0It loaded and according syslog:Nov 11 10:19:47 radius1 genericups[8469]: Startup successfulHowever when I tried: radius1:/etc/nut# upsc [EMAIL PROTECTED] Error: Connection failure: Connection refusedradius1:/etc/nut# I've done this for ttyS0 ttyS1 (there are no more, 2 results in an error at fire up)I also changed the permissions on ttyS1 like radius1:/etc/nut# ls -al /dev/ttyS0crw-rw 1 root nut 4, 64 2005-02-26 07:39 /dev/ttyS0radius1:/etc/nut# Same problem.The only things left here are.1. There is a problem with the serial cable. (I tried 2 of them) How will I know if the cable that came with the UPS is the right one?2. That thing just won't ever work for linux.The new series of MGE are Ellipse ASR I hope someone can confirm these things work as they should with NUT running on debian stable.Many tnxYves Woops forget it!I've noticed a typo in /etc/default/nut # start upsdSTART_UPSD=yed (instead of yes!)Ok, now I gotradius1:/etc/default# upsc [EMAIL PROTECTED] driver.name: genericupsdriver.parameter.port: /dev/ttyS0driver.parameter.upstype: 7driver.version: 2.0.1driver.version.internal: 1.30ups.mfr: CyberPowerups.model: Power99ups.status: OL We move on... ___ Nut-upsuser mailing list Nut-upsuser@lists.alioth.debian.org http://lists.alioth.debian.org/mailman/listinfo/nut-upsuser
Re: [Nut-upsuser] Sweex 1000VA UPS
On 11/11/06, YvesDM [EMAIL PROTECTED] wrote: [...] We move on... Ok, I think it's working. The problem was the serial cable. I had 2, one came with the ups, the other came with another ups. During my tests I changed the original cable with the other one. I never changed it back, thinking they were exactly the same. Now I changed it back and used the serial cabel that came with the ups again.So now we've got:1. On powerline, so status OLradius1:/etc/default# upsc [EMAIL PROTECTED]driver.name: genericupsdriver.parameter.port: /dev/ttyS0driver.parameter.upstype: 7driver.version: 2.0.1driver.version.internal: 1.30ups.mfr: CyberPower ups.model: Power99ups.status: OL2. I pulled the powerplug of the UPS, so OBradius1:/etc/default# upsc [EMAIL PROTECTED]driver.name: genericupsdriver.parameter.port : /dev/ttyS0driver.parameter.upstype: 7driver.version: 2.0.1driver.version.internal: 1.30ups.mfr: CyberPowerups.model: Power99ups.status: OB3. I plugged the powerline back in, again OL :-) radius1:/etc/default# upsc [EMAIL PROTECTED]driver.name: genericupsdriver.parameter.port: /dev/ttyS0driver.parameter.upstype: 7driver.version: 2.0.1driver.version.internal : 1.30ups.mfr: CyberPowerups.model: Power99ups.status: OLradius1:/etc/default# I don't know if the ups will actually shutdown the system now on low battery.upsc ain't showing much information on the battery status here, but as far as I remember I've read somewhere in the docs that this is normal with genericups drivers, not?It's not the right time now to check if the system really is shutdown when the battery is getting low.(live system) I will test this asap and keep this list informed. Many tnxYves ___ Nut-upsuser mailing list Nut-upsuser@lists.alioth.debian.org http://lists.alioth.debian.org/mailman/listinfo/nut-upsuser
Re: [Nut-upsuser] Sweex 1000VA UPS
On 11/11/06, YvesDM [EMAIL PROTECTED] wrote: I don't know if the ups will actually shutdown the system now on low battery. If the status changes to OB LB, then it should shut your system down (assuming everything else is still configured properly). upsc ain't showing much information on the battery status here, but as far as I remember I've read somewhere in the docs that this is normal with genericups drivers, not? Yes, the UPS monitors the battery level internally, and sets the serial pins to a certain state. Because genericups is only looking at the levels of two pins, there are only 4 states that the UPS can report (online vs. on battery, battery OK vs. battery low). There is still the possibility that the UPS can report more status over the serial data lines (including battery charge level, voltages, load, etc.) but it sounds like you have tried a number of different drivers already. It's not the right time now to check if the system really is shutdown when the battery is getting low. (live system) I will test this asap and keep this list informed. Remember that you can also test things with the UPS powering a dummy load, like an old monitor or a lamp. Then, if you simply do not start upsmon, you can discharge the battery completely without shutting down the system which is monitoring the UPS. 'upsc' only needs the driver and 'upsd', as you found out. -- - Charles Lepple ___ Nut-upsuser mailing list Nut-upsuser@lists.alioth.debian.org http://lists.alioth.debian.org/mailman/listinfo/nut-upsuser
Re: [Nut-upsuser] Segmentation fault
Yes, this has been fixed some time ago in the SVN version. -- Peter Magnus Johansson wrote: Hi! I'm running NUT 2.0.4 and occasionally I get a segfault when starting newhidups and if I don't get a segfault when starting newhidups the process starts eating away at my memory. I sat down trying to genererate -D debug info, but just because of that I can't get it to do a segfault. So until I can, has things changed in the development version, that might have fixed this? Network UPS Tools: New USB/HID UPS driver 0.28 (2.0.4) Detected a UPS: American Power Conversion/Back-UPS CS 500 FW:808.q7.I USB FW:q7 Using subdriver: APC/CyberPower HID 0.9 ___ Nut-upsuser mailing list Nut-upsuser@lists.alioth.debian.org http://lists.alioth.debian.org/mailman/listinfo/nut-upsuser ___ Nut-upsuser mailing list Nut-upsuser@lists.alioth.debian.org http://lists.alioth.debian.org/mailman/listinfo/nut-upsuser
Re: [Nut-upsuser] Sweex 1000VA UPS
YvesDM wrote: Ok, I think it's working. The problem was the serial cable. That why I recommended to try it under Windows with the bundled software, to rule out a problem like this. I had 2, one came with the ups, the other came with another ups. During my tests I changed the original cable with the other one. I never changed it back, thinking they were exactly the same. For contact closure type UPSes, they almost never are. Now I changed it back and used the serial cabel that came with the ups again. [...] I don't know if the ups will actually shutdown the system now on low battery. Assuming that the monitoring works (which looks like it), it probably will, as you already found out during earlier tests where your system kept on rebooting when the serial cable was not connected. This is essentially the same as OB LB for upstype=7. upsc ain't showing much information on the battery status here, but as far as I remember I've read somewhere in the docs that this is normal with genericups drivers, not? On Battery and Low Battery are the only two status signals available, so no fancy voltage readings, load percentages and battery charge. If you want that, you need a smarter UPS. It's not the right time now to check if the system really is shutdown when the battery is getting low. That would be a recommendation indeed. See the notes on this in the genericups man page. (live system) I will test this asap and keep this list informed. Good! I'm glad that in the end things worked out for you. Kind regards, Arjen ___ Nut-upsuser mailing list Nut-upsuser@lists.alioth.debian.org http://lists.alioth.debian.org/mailman/listinfo/nut-upsuser
