Tony_YYZ Posted November 14, 2016 Share Posted November 14, 2016 Hi All, I've been running Unraid 5.0.3 for a very long time now. Today I went to connect to one of my SMB shares and notice it wouldn't connect. The Unraid box was not pingable either. I turned on the monitor at the console and saw it was sitting at my BIOS' "You're settings have been reset" screen. I went into the BIOS, set the time, ensured the boot sequence was correct, saved and restarted. Each time the server boots into Unraid, it randomly freezes up or the whole system restarts. I've tried the following: 1. Performed a Checkdisk on my USB key in another PC. Windows found errors and corrected them. - NO CHANGE, ISSUES STILL PERSIST 2. Upgraded from 5.0.3 to 6.2.4. - NO CHANGES, ISSUES STILL PERSIST 3. Tried running MemTest, it froze after running for 24 minutes - NO CHANGES, ISSUES STILL PERSIST I've run out of options to try. I keep trying to run a Parity Check upon getting back into the WebUI but the system keeps failing before it can even reach 10%. I have 4 data disks, 1 parity and 1 cache disk. I can't capture a System Log because the system freezes up and becomes completely non-responsive. I did keep refreshing the log and eventually just copied the info to a text file (see attached). When the system freezes, sometimes it locks up at the login prompt, other times it starts rapidly cycling strings of text that I can't make out on the display. The only thing I was able to read was something along the lines of "i2c_core ahci libahci r8168". Sometimes the display showed horizontal lines in rainbow colours. I have my Unraid server attached to an APC UPS at all times. I did not experience any power outages recently. My specs are as follows: -MSI H77MA-G43 Motherboard (w/onboard NIC and Graphics) -Intel i3-2120 CPU -8GB GSkill Sniper 1.25v RAM -Antec Neo-ECO 450w PSU I'm hoping someone can help shed some light on what's wrong with it. I've had it running for years without issue. I appreciate any and all responses. Thank you. unraid_log_nov13.txt Quote Link to comment
JorgeB Posted November 14, 2016 Share Posted November 14, 2016 If it froze during memtest you have a hardware issue, it could be ram, board, power supply, etc Quote Link to comment
itimpi Posted November 14, 2016 Share Posted November 14, 2016 It could be something like the CPU fan not working properly thus causing the system to be shutdown due to the CPU getting too hot and burning out. Quote Link to comment
Tony_YYZ Posted November 14, 2016 Author Share Posted November 14, 2016 If it froze during memtest you have a hardware issue, it could be ram, board, power supply, etc That's my guess as well. Any tips for trying to narrow it down? Quote Link to comment
Tony_YYZ Posted November 14, 2016 Author Share Posted November 14, 2016 It could be something like the CPU fan not working properly thus causing the system to be shutdown due to the CPU getting too hot and burning out. After searching this forum to see if anyone else was experiencing a similar issue, I did happen to check if the CPU fan was working. It is working fine and the CPU temp is around 40-42 Celsius during the time I was monitoring it. Quote Link to comment
JorgeB Posted November 14, 2016 Share Posted November 14, 2016 If it froze during memtest you have a hardware issue, it could be ram, board, power supply, etc That's my guess as well. Any tips for trying to narrow it down? Start replacing parts, RAM should be the easiest if you have more than one DIMM, try one at a time. Quote Link to comment
Zonediver Posted November 14, 2016 Share Posted November 14, 2016 Powersupply or RAM - the most responsible Parts for this Problem The Board is from 2012 - should be ok but could be the problem too. Quote Link to comment
Tony_YYZ Posted November 14, 2016 Author Share Posted November 14, 2016 If it froze during memtest you have a hardware issue, it could be ram, board, power supply, etc That's my guess as well. Any tips for trying to narrow it down? Start replacing parts, RAM should be the easiest if you have more than one DIMM, try one at a time. I'll definitely start with that. Thank you. Quote Link to comment
Tony_YYZ Posted November 14, 2016 Author Share Posted November 14, 2016 Powersupply or RAM - the most responsible Parts for this Problem The Board is from 2012 - should be ok but could be the problem too. I really hope it's not the motherboard. That would be the most pain in the ass part to replace. But knowing my luck.... I'm going to start testing RAM now. I have to research a new power supply in the meantime as well. Not sure if my current PSU is overkill or just right at 450W ~87% efficiency. Quote Link to comment
Zonediver Posted November 14, 2016 Share Posted November 14, 2016 Powersupply or RAM - the most responsible Parts for this Problem The Board is from 2012 - should be ok but could be the problem too. I really hope it's not the motherboard. That would be the most pain in the ass part to replace. But knowing my luck.... I'm going to start testing RAM now. I have to research a new power supply in the meantime as well. Not sure if my current PSU is overkill or just right at 450W ~87% efficiency. The Question is: How many Disks are in your Server and have you already measured the Power-Input (with a Power-Meter)? As orientation, my Sys needs a peek power of 200W during Boot - measured over a Power-Meter on the 220V-Input. Take care about the SATA-Connectors on the PSU - thats the most important point. Quote Link to comment
Frank1940 Posted November 14, 2016 Share Posted November 14, 2016 Powersupply or RAM - the most responsible Parts for this Problem The Board is from 2012 - should be ok but could be the problem too. I really hope it's not the motherboard. That would be the most pain in the ass part to replace. But knowing my luck.... I'm going to start testing RAM now. I have to research a new power supply in the meantime as well. Not sure if my current PSU is overkill or just right at 450W ~87% efficiency. The Question is: How many Disks are in your Server and have you already measured the Power-Input (with a Power-Meter)? As orientation, my Sys needs a peek power of 200W during Boot - measured over a Power-Meter on the 220V-Input. Take care about the SATA-Connectors on the PSU - thats the most important point. It is the +12V buss that is the worry. You want a single rail supply for unRAID because the only use for 12V in a server is the disk drives. When looking at your old supply, you have to be very careful about the ampere ratings on the various voltage busses. I have seen 450W supplies that when you total up the wattage for the various busses comes out well over 500W! Plus, if you do exceed the wattage any of the busses, often the entire supply will shutdown. When (and if) you start to look at PS, look for quality and reliability (negative user reviews are a good source of information here) rather than price. A PS is usually the only piece of hardware that you can use in your next major upgrade! Quote Link to comment
Tony_YYZ Posted November 14, 2016 Author Share Posted November 14, 2016 Powersupply or RAM - the most responsible Parts for this Problem The Board is from 2012 - should be ok but could be the problem too. I really hope it's not the motherboard. That would be the most pain in the ass part to replace. But knowing my luck.... I'm going to start testing RAM now. I have to research a new power supply in the meantime as well. Not sure if my current PSU is overkill or just right at 450W ~87% efficiency. The Question is: How many Disks are in your Server and have you already measured the Power-Input (with a Power-Meter)? As orientation, my Sys needs a peek power of 200W during Boot - measured over a Power-Meter on the 220V-Input. Take care about the SATA-Connectors on the PSU - thats the most important point. I have 3x Seagate 1TB 7200RPM disks, 2x Western Digital Red 3TB disks and 1x Western Digital Blue for the Cache Disk. I've never measured the input before. I'm eyeing the SeaSonic G Series SSR-550RM PSU as replacement unit. It's one of the more economical solutions available to me here in Canada with a single rail 12V. Quote Link to comment
Tony_YYZ Posted November 14, 2016 Author Share Posted November 14, 2016 Powersupply or RAM - the most responsible Parts for this Problem The Board is from 2012 - should be ok but could be the problem too. I really hope it's not the motherboard. That would be the most pain in the ass part to replace. But knowing my luck.... I'm going to start testing RAM now. I have to research a new power supply in the meantime as well. Not sure if my current PSU is overkill or just right at 450W ~87% efficiency. The Question is: How many Disks are in your Server and have you already measured the Power-Input (with a Power-Meter)? As orientation, my Sys needs a peek power of 200W during Boot - measured over a Power-Meter on the 220V-Input. Take care about the SATA-Connectors on the PSU - thats the most important point. It is the +12V buss that is the worry. You want a single rail supply for unRAID because the only use for 12V in a server is the disk drives. When looking at your old supply, you have to be very careful about the ampere ratings on the various voltage busses. I have seen 450W supplies that when you total up the wattage for the various busses comes out well over 500W! Plus, if you do exceed the wattage any of the busses, often the entire supply will shutdown. When (and if) you start to look at PS, look for quality and reliability (negative user reviews are a good source of information here) rather than price. A PS is usually the only piece of hardware that you can use in your next major upgrade! I'm eyeing the SeaSonic G Series SSR-550RM PSU as replacement unit. It's one of the more economical solutions available to me here in Canada with a single rail 12V while being a good quality unit from what I can ascertain. As it stands, I pulled one of my DIMM units (4GB of 8GB total) and my server has been online all day so far. The UI is faster, my APC UPS metrics within Unraid shows the server using less power than it did last night which is odd based on the UPS load rating. My parity check is about 1.5hrs away from completing. Once that is done, I'll mount the disks and attempt normal operations again and see how it behaves. I'll start spec'ing out replacement RAM modules as well since now that I can run VMs within Unraid, I'll want to take advantage of that. Quote Link to comment
Zonediver Posted November 14, 2016 Share Posted November 14, 2016 Good that it was the RAM and not the Mainboard Quote Link to comment
Tony_YYZ Posted November 14, 2016 Author Share Posted November 14, 2016 Good that it was the RAM and not the Mainboard Fingers crossed that it is indeed the RAM module. Quote Link to comment
JonathanM Posted November 14, 2016 Share Posted November 14, 2016 Good that it was the RAM and not the Mainboard Fingers crossed that it is indeed the RAM module. The other RAM module may not be bad. It's quite possible that your motherboard can no longer successfully power 2 sticks of RAM simultaneously. I'd swap sticks and make SURE that the other stick is truly bad before coming to a conclusion. Quote Link to comment
Zonediver Posted November 15, 2016 Share Posted November 15, 2016 Good idea - always test both RAMs to see, if it "is" the RAM or the RAM-Slot on the Mainboard. Quote Link to comment
tokra Posted December 6, 2016 Share Posted December 6, 2016 I am also experiencing sporadical freezing of system, after some longer time -several hours. Running 6.2.4. I did check flash but its okay, today i'm gonna check RAMs under memtest86. i dont know what else could be wrong. I ungraded to completely new HW, but i also experienced UNRAID freezing several times, during last month on previous HW with 6.2.1, which is strange, both HW wrong ? Could be corrupted UNRAID system on flash ? I memtest shows no error i might reformat flash and setup brand new UNRAID again. :-( Quote Link to comment
Zonediver Posted December 6, 2016 Share Posted December 6, 2016 I am also experiencing sporadical freezing of system, after some longer time -several hours. Running 6.2.4. I did check flash but its okay, today i'm gonna check RAMs under memtest86. i dont know what else could be wrong. I ungraded to completely new HW, but i also experienced UNRAID freezing several times, during last month on previous HW with 6.2.1, which is strange, both HW wrong ? Could be corrupted UNRAID system on flash ? I memtest shows no error i might reformat flash and setup brand new UNRAID again. :-( What kind of NIC is in your machine? I had several strange things with unraid and then i changed the NIC from Realtek-Onboard to Intel Pro-1000 and all strange things are gone, so maybe the NIC is the problem? Quote Link to comment
tokra Posted December 7, 2016 Share Posted December 7, 2016 I am also experiencing sporadical freezing of system, after some longer time -several hours. Running 6.2.4. I did check flash but its okay, today i'm gonna check RAMs under memtest86. i dont know what else could be wrong. I ungraded to completely new HW, but i also experienced UNRAID freezing several times, during last month on previous HW with 6.2.1, which is strange, both HW wrong ? Could be corrupted UNRAID system on flash ? I memtest shows no error i might reformat flash and setup brand new UNRAID again. :-( What kind of NIC is in your machine? I had several strange things with unraid and then i changed the NIC from Realtek-Onboard to Intel Pro-1000 and all strange things are gone, so maybe the NIC is the problem? Hi, I am using one which is onboard: Qualcomm® Atheros® AR8171 http://www.asrock.com/mb/AMD/FM2A88M%20Extreme4+%20R2.0/ Why do you think NIC is the problem ? Quote Link to comment
Frank1940 Posted December 7, 2016 Share Posted December 7, 2016 Zonediver: One thing. You should start your own thread so that people would recognize that there is (most probably) a whole new problem to be addressed. It is to your benefit to do this to get as many folks as possible to see your issues. Second thing, after reading that you had this issue with other hardware and an earlier version of the unRAID software, you should also post up a diagnostics file. I suspect that you have an out-of-date plugin that is giving you a problem. Quote Link to comment
tokra Posted December 7, 2016 Share Posted December 7, 2016 Here is diagnostic tower-diagnostics-20161206-0532.zip Quote Link to comment
tokra Posted December 8, 2016 Share Posted December 8, 2016 So I have checked memory in memtest86 and no error. So i reformated flash and put brand new clean unraid. Installed dockers. Unraid was running over night, when i woke up i found out it again freezed. :-( Quote Link to comment
Frank1940 Posted December 8, 2016 Share Posted December 8, 2016 So I have checked memory in memtest86 and no error. So i reformated flash and put brand new clean unraid. Installed dockers. Unraid was running over night, when i woke up i found out it again freezed. :-( How long did you let the memtst run? Twenty-four hours is considered to be the minimum. You said you replaced the Hardware. Did you reuse the Power Supply or any other components except the disk drives? ( I seem to recall that the PS has been the culprit in a couple of cases with similar symptoms.) Quote Link to comment
Zonediver Posted December 8, 2016 Share Posted December 8, 2016 Zonediver: One thing. You should start your own thread so that people would recognize that there is (most probably) a whole new problem to be addressed. It is to your benefit to do this to get as many folks as possible to see your issues. Second thing, after reading that you had this issue with other hardware and an earlier version of the unRAID software, you should also post up a diagnostics file. I suspect that you have an out-of-date plugin that is giving you a problem. ? I dont have problems with unraid - not any more so my post was just a guess, what might be the problem Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.