Troubleshooting

• 

  THE DANCE - Physical hardware troubleshooting
  • dbct  

  1
  1
  Votes
  1
  Posts
  3849
  Views

  

  The OS is software but this does not mean that we stop there, we will try to help you also with some common hardware troubleshooting and some tips and tricks How to get a new rig running step by step: Let's start simple: remove all cards but one. Put that card in the slot on the board and make sure that PiMP boots and mines. Then, turn off your rig and put that card on a riser in the first slot and turn it on. Make sure that the GPU boots and mines while on the riser. If this runs, then gradually add more cards on risers one at a time. One GPU is not detected, stops mining or responding, or you suspect you have a riser failing: Great now you can turn off the rig and connect a known working card to that riser and start the rig up again. Did that fix the problem? Then you know something might be wrong with the GPU, not the riser. If this did not fix the problem, try a new riser and see if that works. If it does, then that riser is probably broken. Also, don’t forget to re-check and re-seat all power connectors for both the GPUs and the risers. Also, if you are using powered risers if the power connector for that riser has a good connection. Protip : Avoid using more than 2 risers per power cable string coming from your psu(Power supply unit)!! this to avoid voltage dips or in worst case a fire GPUs are not detected or miner does not want to start: Almost all hardware troubleshooting information is found in the PiMP Troubleshooter / helpme URL. Take a look and see if you can identify your problem. Hardware issues are usually found in the first section or the GPU INFO section of the Troubleshooter. If for example you see a note saying insufficient power in the nvidia-sm section then please recheck power connectors that come from your psu(Power supply unit) to your card. If you cannot identify your problem after looking at the Troubleshooter, search the forums, and if you cannot find your problem there, go to the PIMP Discord chat and post your helpme there to see if any other miners can help you. When you had contact with pimp support and you have send them a helpme they can see most of the time what is going on. Basic hardware troubleshooting (aka the Hardware Dance!) If the OS does not see your cards directly or you have different cards in your rig, start swapping. Only change one thing at a time. The PiMP shuffle is to move cards around in your rig one card at a time and boot it up every time to see when all are detected. This also counts for the cases that you suspect something is wrong on the hardware side, a swap or shuffle with a known good is always a good place to start and get yourself back to mining the quickest way. This technique works for troubleshooting all types of things, not just miners or computer systems.
• 

  What to do if your rig is hard locking up.
  • melt  

  1
  1
  Votes
  1
  Posts
  1954
  Views

  

  Here is some information you can use when your rig is locking up (the worst case scenario) which does not happen often. 99.999% hardware problem, but PiMP OS has tools to fix it. First, some information: How miner.farm talks to the rigs: Miner.farm sends commands to the rig via the agent process, when it checks in. Since we do not store any IPs and ports and passwords, the system is completely secure. However, there is just no way for us to send a command to the rig otherwise, if the agent service is not running. The agent service itself (minerfarm_svc) is managed by systemd, the linux system process manager. This is the modern way of managing system processes on Ubuntu based systems (and most modern distros of Linux.) If the agent stops, systemd will try to restart it within limits, automatically. The end result is, if your agent is stopped, your rig is likely down or has a problem that requires your intervention. First things first Try to VNC or RDP to the rig. Instructions are in the Getting Started forum category post here: https://forum.getpimp.org/topic/16/remote-access-using-rdp-vnc Try to SSH into the rig. If SSH is not answering, then it is likely the system is hard locked or not powered on properly. SSH is fairly resilient, even if a system is corrupted fairly badly. SSH example using MobaXterm for Windows found in the video here on the Quickstart Guide: https://forum.getpimp.org/topic/15/quickstart-guide Next thing would be, if you have a monitor connected, to go to the rig and see what is on the local display. If you booted headless, it is likely that the rig will not output video, so you could skip this part. If you cannot get to the rig using these methods, try connecting a display so you can see what is going on (optional) and power cycle the rig. After a few moments it should be accessible using the methods above. At this time you would want to troubleshoot right away.... using such things like: nano startup.sh and block the starting of the miner with a # - see if the rig runs for a while without mining. If this is the case, then you know mining activity is part of the cause. Run the helpme and view the report. You can skip right to the GPU Info section if you like. This is where you can see errors. Also look at the dmesg log. You can see the names of the commands there. Look in /var/log for the miner logs and for the agent.log. This might give a clue. Watch the troubleshooting video and read the guides for troubleshooting tools and hardware dance here: https://forum.getpimp.org/category/6/how-to-troubleshoot-your-rig Physical remote reboot options: Z-wave option: If you have a smart home alarm system you can use a Z-wave or Zigbee controller or Vera, Google Home, etc. https://www.amazon.com/gp/product/B0013V58HU/ref=oh_aui_search_detailpage?ie=UTF8&psc=1 Wifi smart plug options are available as well. Auto reboot option: This is not recommended. (Band-aid fix) nano /etc/crontab if you want examples on auto rebooting on a scheduled interval - perhaps that's another option for you - but again - bandaid compared to troubleshooting the cause of the problem (before it burns the rig up) - You should really look for the cause and resolve it instead of just kicking the box over and over. The box will not last long in this setup.
• 

  How to get help - The troubleshooting tools
  • CryptoNews  

  2
  0
  Votes
  2
  Posts
  5444
  Views

  

  Note, you can also run the helpme report remotely using miner.farm. This will go to the rig, run helpme, and if it completes successfully, provide you the URL right in the Host Hardware Info section on your Farm Admin > Agent View screen. For how-to guide: See this article in Miner.farm Support forum category: https://forum.getpimp.org/topic/5910/how-to-run-the-helpme-pimp-os-diagnostics-remotely-from-miner-farm Happy hunting those problems, miners!
• 

  Things to try when gputool --config throws errors or wont set the clocks
  • melt  

  1
  1
  Votes
  1
  Posts
  1855
  Views

  

  "they mine just fine. it just wont let me adjust the clock speeds" You may have already tried this, but this is one solution that worked for me, start off with one riser and a 1060 gpu plugged in, then see if by running “gputool status” then error goes away. If the error go away, then plug the next 1060 gpu and check again. If all works' then plug in your 580's one at a time, and check again. Another option that worked for me is, If you have a spare ssd, and time of course, might be worth re-imaging to 2.8.6, once key validated and PiMP OS running, don’t update the pimp OS version to latest or run “apt update” nor “apt upgrade”, keep everything to default, then run “gputool —status”, and see if all gpu’s and their current settings are displayed, if not try with again just one 1060 gpu plugged in. Hope this helps, and let us known how you get on. My next thought as a fellow mixed rig user, is are you using in the BIOS settings Gen1 or Gen2, reason being is I had some issues using Gen2 and known for my mixed 1070/570 rig it prefered Gen1. Lastly, once the error is resolved, might be worth looking at phoenixminer.pcfg, as it has 0.65% dev fee, and although it's a "copy" of claymore.dual in many ways, has some performance improvements, and allows miner.farm to display more of the states apart from Power watts, which is being looked at by the dev team. Also, once "gputool --status" displays the AMD GPU states, might be worth using the default "gputool --config" values for the AMD GPU", just in case your own modified values, cause an issue with gputool, might not but trying to cover all bases. Again let us know how you get on, and your not alone :smiley: You could also try removing the current "gpu-config.json" file by running command "rm gpu-config.json", and then re-creating the file by running command "gputool --config". Then once the new "gpu-config.json" file is created, commit it by running again "gputool --config", and see whether the default values for your 1060's are applied without error's like they currently being displayed without any errors for your 3 580's. gputool --reset will also regenerate a new gpu-config.json. Credit: Louis, PIMP OS Support Team
• 

  Linux Network Troubleshooting tools available in the command line
  • melt  

  1
  0
  Votes
  1
  Posts
  3077
  Views

  

  Overview Here is a list of common network troubleshooting tools that you can use in most linux terminals (debian/ubuntu/redhat/fedora distros). IFCONFIG command ifconfig (interface configurator) command is use to initialize an interface, assign IP Address to interface and enable or disable interface on demand. With this command you can view IP Address and Hardware / MAC address assign to interface and also MTU (Maximum transmission unit) size. # ifconfig eth0 Link encap:Ethernet HWaddr 00:0C:29:28:FD:4C inet addr:192.168.50.2 Bcast:192.168.50.255 Mask:255.255.255.0 inet6 addr: fe80::20c:29ff:fe28:fd4c/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:6093 errors:0 dropped:0 overruns:0 frame:0 TX packets:4824 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:6125302 (5.8 MiB) TX bytes:536966 (524.3 KiB) Interrupt:18 Base address:0x2000 lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 inet6 addr: ::1/128 Scope:Host UP LOOPBACK RUNNING MTU:16436 Metric:1 RX packets:8 errors:0 dropped:0 overruns:0 frame:0 TX packets:8 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:480 (480.0 b) TX bytes:480 (480.0 b) ifconfig with interface (eth0) command only shows specific interface details like IP Address, MAC Address etc. with -a options will display all available interface details if it is disable also. # ifconfig eth0 eth0 Link encap:Ethernet HWaddr 00:0C:29:28:FD:4C inet addr:192.168.50.2 Bcast:192.168.50.255 Mask:255.255.255.0 inet6 addr: fe80::20c:29ff:fe28:fd4c/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:6119 errors:0 dropped:0 overruns:0 frame:0 TX packets:4841 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:6127464 (5.8 MiB) TX bytes:539648 (527.0 KiB) Interrupt:18 Base address:0x2000 Assigning IP Address and Gateway Assigning an IP Address and Gateway to interface on the fly. The setting will be removed in case of system reboot. # ifconfig eth0 192.168.50.5 netmask 255.255.255.0 Enable or Disable Specific Interface To enable or disable specific Interface, we use example command as follows. Enable eth0 # ifup eth0 Disable eth0 # ifdown eth0 Setting MTU Size By default MTU size is 1500. We can set required MTU size with below command. Replace XXXX with size. # ifconfig eth0 mtu XXXX Set Interface in Promiscuous mode Network interface only received packets belongs to that particular NIC. If you put interface in promiscuous mode it will received all the packets. This is very useful to capture packets and analyze later. For this you may require superuser access. # ifconfig eth0 - promisc 2. PING Command PING (Packet INternet Groper) command is the best way to test connectivity between two nodes. Whether it is Local Area Network (LAN) or Wide Area Network (WAN). Ping use ICMP (Internet Control Message Protocol) to communicate to other devices. You can ping host name of ip address using below command. # ping 4.2.2.2 PING 4.2.2.2 (4.2.2.2) 56(84) bytes of data. 64 bytes from 4.2.2.2: icmp_seq=1 ttl=44 time=203 ms 64 bytes from 4.2.2.2: icmp_seq=2 ttl=44 time=201 ms 64 bytes from 4.2.2.2: icmp_seq=3 ttl=44 time=201 ms OR # ping www.example.com PING example.com (50.64.202.120) 56(84) bytes of data. 64 bytes from 50.64.202.120: icmp_seq=1 ttl=47 time=284 ms 64 bytes from 50.64.202.120: icmp_seq=2 ttl=47 time=287 ms 64 bytes from 50.64.202.120: icmp_seq=3 ttl=47 time=285 ms In Linux ping command keep executing until you interrupt. Ping with -c option exit after N number of request (success or error respond). # ping -c 5 www.example.com PING example.com (50.64.202.120) 56(84) bytes of data. 64 bytes from 50.64.202.120: icmp_seq=1 ttl=47 time=285 ms 64 bytes from 50.64.202.120: icmp_seq=2 ttl=47 time=285 ms 64 bytes from 50.64.202.120: icmp_seq=3 ttl=47 time=285 ms 64 bytes from 50.64.202.120: icmp_seq=4 ttl=47 time=285 ms 64 bytes from 50.64.202.120: icmp_seq=5 ttl=47 time=285 ms --- example.com ping statistics --- 5 packets transmitted, 5 received, 0% packet loss, time 4295ms rtt min/avg/max/mdev = 285.062/285.324/285.406/0.599 ms 3. TRACEROUTE Command traceroute is a network troubleshooting utility which shows number of hops taken to reach destination also determine packets traveling path. Below we are tracing route to global DNS server IP Address and able to reach destination also shows path of that packet is traveling. # traceroute 4.2.2.2 traceroute to 4.2.2.2 (4.2.2.2), 30 hops max, 60 byte packets 1 192.168.50.1 (192.168.50.1) 0.217 ms 0.624 ms 0.133 ms 2 227.18.106.27.mysipl.com (27.106.18.227) 2.343 ms 1.910 ms 1.799 ms 3 221-231-119-111.mysipl.com (111.119.231.221) 4.334 ms 4.001 ms 5.619 ms 4 10.0.0.5 (10.0.0.5) 5.386 ms 6.490 ms 6.224 ms 5 gi0-0-0.dgw1.bom2.pacific.net.in (203.123.129.25) 7.798 ms 7.614 ms 7.378 ms 6 115.113.165.49.static-mumbai.vsnl.net.in (115.113.165.49) 10.852 ms 5.389 ms 4.322 ms 7 ix-0-100.tcore1.MLV-Mumbai.as6453.net (180.87.38.5) 5.836 ms 5.590 ms 5.503 ms 8 if-9-5.tcore1.WYN-Marseille.as6453.net (80.231.217.17) 216.909 ms 198.864 ms 201.737 ms 9 if-2-2.tcore2.WYN-Marseille.as6453.net (80.231.217.2) 203.305 ms 203.141 ms 202.888 ms 10 if-5-2.tcore1.WV6-Madrid.as6453.net (80.231.200.6) 200.552 ms 202.463 ms 202.222 ms 11 if-8-2.tcore2.SV8-Highbridge.as6453.net (80.231.91.26) 205.446 ms 215.885 ms 202.867 ms 12 if-2-2.tcore1.SV8-Highbridge.as6453.net (80.231.139.2) 202.675 ms 201.540 ms 203.972 ms 13 if-6-2.tcore1.NJY-Newark.as6453.net (80.231.138.18) 203.732 ms 203.496 ms 202.951 ms 14 if-2-2.tcore2.NJY-Newark.as6453.net (66.198.70.2) 203.858 ms 203.373 ms 203.208 ms 15 66.198.111.26 (66.198.111.26) 201.093 ms 63.243.128.25 (63.243.128.25) 206.597 ms 66.198.111.26 (66.198.111.26) 204.178 ms 16 ae9.edge1.NewYork.Level3.net (4.68.62.185) 205.960 ms 205.740 ms 205.487 ms 17 vlan51.ebr1.NewYork2.Level3.net (4.69.138.222) 203.867 ms vlan52.ebr2.NewYork2.Level3.net (4.69.138.254) 202.850 ms vlan51.ebr1.NewYork2.Level3.net (4.69.138.222) 202.351 ms 18 ae-6-6.ebr2.NewYork1.Level3.net (4.69.141.21) 201.771 ms 201.185 ms 201.120 ms 19 ae-81-81.csw3.NewYork1.Level3.net (4.69.134.74) 202.407 ms 201.479 ms ae-92-92.csw4.NewYork1.Level3.net (4.69.148.46) 208.145 ms 20 ae-2-70.edge2.NewYork1.Level3.net (4.69.155.80) 200.572 ms ae-4-90.edge2.NewYork1.Level3.net (4.69.155.208) 200.402 ms ae-1-60.edge2.NewYork1.Level3.net (4.69.155.16) 203.573 ms 21 b.resolvers.Level3.net (4.2.2.2) 199.725 ms 199.190 ms 202.488 ms 4. NETSTAT Command Netstat (Network Statistic) command display connection info, routing table information etc. To displays routing table information use option as -r. # netstat -r Kernel IP routing table Destination Gateway Genmask Flags MSS Window irtt Iface 192.168.50.0 * 255.255.255.0 U 0 0 0 eth0 link-local * 255.255.0.0 U 0 0 0 eth0 default 192.168.50.1 0.0.0.0 UG 0 0 0 eth0 For more examples of Netstat Command, please read our earlier article on 20 Netstat Command Examples in Linux. 5. DIG Command Dig (domain information groper) query DNS related information like A Record, CNAME, MX Record etc. This command mainly use to troubleshoot DNS related query. # dig www.example.com; <<>> DiG 9.8.2rc1-RedHat-9.8.2-0.10.rc1.el6 <<>> www.example.com ;; global options: +cmd ;; Got answer: ;; ->>HEADER< For more examples of Dig Command, please read the article on 10 Linux Dig Commands to Query DNS. 6. NSLOOKUP Command nslookup command also use to find out DNS related query. The following examples shows A Record (IP Address) of example.com. # nslookup www.example.com Server: 4.2.2.2 Address: 4.2.2.2#53 Non-authoritative answer: www.example.com canonical name = example.com. Name: example.com Address: 50.64.202.120 For more NSLOOKUP Command, read the article on 8 Linux Nslookup Command Examples. ``` # 7. ROUTE Command route command also shows and manipulate ip routing table. To see default routing table in Linux, type the following command. ``` # route Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface 192.168.50.0 * 255.255.255.0 U 0 0 0 eth0 link-local * 255.255.0.0 U 1002 0 0 eth0 default 192.168.50.1 0.0.0.0 UG 0 0 0 eth0 ``` Adding, deleting routes and default Gateway with following commands. ## Route Adding `# route add -net 10.10.10.0/24 gw 192.168.0.1` Route Deleting `# route del -net 10.10.10.0/24 gw 192.168.0.1` Adding default Gateway `# route add default gw 192.168.0.1` # 8. HOST Command host command to find name to IP or IP to name in IPv4 or IPv6 and also query DNS records. ``` # host www.google.com www.google.com has address 173.194.38.180 www.google.com has address 173.194.38.176 www.google.com has address 173.194.38.177 www.google.com has address 173.194.38.178 www.google.com has address 173.194.38.179 www.google.com has IPv6 address 2404:6800:4003:802::1014 ``` Using -t option we can find out DNS Resource Records like CNAME, NS, MX, SOA etc. ``` # host -t CNAME www.redhat.com www.redhat.com is an alias for wildcard.redhat.com.edgekey.net. ``` # 9. ARP Command ARP (Address Resolution Protocol) is useful to view / add the contents of the kernel’s ARP tables. To see default table use the command as. ``` # arp -e Address HWtype HWaddress Flags Mask Iface 192.168.50.1 ether 00:50:56:c0:00:08 C eth0 ``` # 10. ETHTOOL Command ethtool is a replacement of mii-tool. It is to view, setting speed and duplex of your Network Interface Card (NIC). You can set duplex permanently in /etc/sysconfig/network-scripts/ifcfg-eth0 with ETHTOOL_OPTS variable. ``` # ethtool eth0 Settings for eth0: Current message level: 0x00000007 (7) Link detected: yes ``` # 11. IWCONFIG Command iwconfig command in Linux is use to configure a wireless network interface. You can see and set the basic Wi-Fi details like SSID channel and encryption. You can refer man page of iwconfig to know more. `# iwconfig [interface]` # 12. HOSTNAME Command hostname is to identify in a network. Execute hostname command to see the hostname of your box. You can set hostname permanently in /etc/sysconfig/network. Need to reboot box once set a proper hostname. ``` # hostname example.com ```
• 

  How to take screenshots in Windows
  • melt  

  1
  0
  Votes
  1
  Posts
  2363
  Views

  

  Screenshots: A picture is worth 1000 words. Even from Windows. ….especially for worldwide pimps who know many languages, it makes life easy to show a screen to a fellow PiMP trying to help you. Just like we make the guides with! You can use PrntScr and Alt-PrntScr key to paste to clipboard, but it’s a whole lot nicer to be able to crop, save, and instantly cloud-post. For this, the freeware Lightshot seems to work good. You can download it from their Homepage: https://app.prntscr.com/en/index.html Another great browser-based one is really really easy to use and multi-platform: https://snag.gy Update: here are some other good image hosting places and utilities: https://imgur.com https://gyazo.com Enjoy! And thank me below :) Note: This software is not made by PiMP and they did not pay to have us recommend it: it is simply the best and/or one of the ones that we know of.
• 

  How and when to reimage an existing rig
  • dbct  

  1
  0
  Votes
  1
  Posts
  4575
  Views

  

  For major upgrades to PiMP, beyond a pimpup, you can re-image your rig. This is done when the OS has a critical update that can’t or is unsafe to be done via pimpup, for example kernel updates or other major OS changes. In order for you to do this safely, please save your confs and pcfg or just set them up again after. After this is done shut down and download the new and fresh image from your account on the getpimp.org site Now unplug the drive (ssd or hdd) and connect it to your img pc. What you need to do is follow the same image process as you done befor getting the img on the disk. If you have verified this key recently, please give our servers a bit of time to make your key ready to validate again, this should be a hour after you verified last time at the least. Once the image process is complete, put the disk back into your rig and validate your rig again with the new image. Everything should be in order and if you saved the confs and pcfgs feel free to put them back. Things should now work the same or even better than before, if this is not the case please read the forum post with the new bugs and workarounds. If this does not have the solution for you then jump in the #pimp discord and ask around. Examples of Disk Corruption Any Busybox or initramfs message at boot time means you will have to reimage. How to backup your configs: Cloud pasting you can just paste them to the cloud. SCP to other rig you can copy to another rig.
• 

  PiMP Outage and Emergency Procedures
  testers guide emergency in case of emergency • • melt  

  1
  0
  Votes
  1
  Posts
  1371
  Views

  

  Overview This document will serve to outline the process to follow if there is an outage or other pimp emergency. Code Green (Support Issues) If: Users fighting or spamming the chat A Customer mentions they might post something negative about PIMP. Then: Highlight @Support in Discord. Support: Highlight @Staff in the Support channel to alert Staff. Code Yellow (Moderate) If: miner.farm slow getpimp.org slow problem with a pimpup new issue reported by more than 3 users Then: Highlight @melt on discord immediately. If no answer on discord within 5 minutes, highlight @staff. Code Red (Outage/Severe) If: miner.farm down getpimp.org down problem with checkout on any site Then: Highlight @staff on discord immediately DM @melt on discord immediately. If no answer on discord within 5 minutes, text melt immediately. (@staff only)
• 

  Articles on Image IMAGING vs CLONING
  • Paz  

  1
  1
  Votes
  1
  Posts
  3770
  Views

  

  This is more so I dont lose the pages. I along with many others consistently have problems with key validation, and flashing of the HD images. While the team is working hard on resolving this, it seems to be the same issue over and over again. People are used to "installing" software which means a program puts files on your operating system and the installers these days have checks to make sure everything got to where it should. Imaging is not the same thing. You take an IMG file and copy the data bit by bit to a destination drive. Based on how this works the utility has no way to know or determine if the source bit and destination bit are identical or correct. Assumption #1: Because my image loads linux, it should be fine Assumption #2: Because the key validator at least ran (and failed) the image flashed fine Apparently the data suggest this is not the case. If a few bits fail in random places, only certain things will not work properly. The team is trying very hard to come up with a new process that takes this uncertainty away. But for the meantime, PIMP DOES WORK and 1000s of users are doing this fine day to do. Dont be discouraged, have some faith. It's frustrating to do this over and over again without any luck, but that doesnt mean Pimp is broken. Here are a few links which I found today which might help. This guru knows HDDs inside and out and describes the difference between IMAGING and CLONING and what happens when you unknowingly write to a bad sector. Even brand new drives contain bad sectors, and truth be told, SSDs are even worse for this. SSD technology is known for static errors, so they implemented types of error correction which CANNOT be used in the img flashing process. http://www.hddoracle.com/viewtopic.php?f=117&t=1410 This article discusses a utility that helps you determine there are bad sectors and mark them to not be used. http://www.hddoracle.com/viewtopic.php?f=30&t=1183 I havent been able to test this yet, but will do so this week. If anyone beats me to the punch, please update here.
• 

  Top Known Bugs, Problems, and Questions
  • melt  

  2
  1
  Votes
  2
  Posts
  6853
  Views

  

  xmrig-cpu libssl error on 2.9.3+ Summary: error /opt/miners/xmrig-cpu/xmrig-cpu: error while loading shared libraries: libssl.so.1.1: cannot open shared object file: No such file or directory Status: PiMP 2.9+ will contain this package. Workaround: apt install -y libssl1.1 is the fix Reason: They are using a newer version of the ssl library than what originally shipped with ubuntu.
• 

  SOLVED What is a Sick or Dead gpu? & how to fix
  • dasgutya  

  5
  0
  Votes
  5
  Posts
  7353
  Views

  S

  For me, the dead GPU issue was caused by poor GPU spacing. Nothing indicated overheating but I separated my GPU's to give them twice as much space and haven't had the problem again. When I did this, my hash rate also jumped up so I apparently wasn't given them enough breathing room.
• 

  Warnings and Error messages that can be safely ignored
  • melt  

  1
  0
  Votes
  1
  Posts
  3213
  Views

  

  Error : can't find command 'gfxmode' This is just a warning from an old boot configuration file. It does not cause any issues. tsc_firmware: This is just about CPU microcode firmware and does not affect mining at all. Error in creating .profile: This happened on older versions of PIMP on opening the GUI.
• 

  SOLVED i’m getting Cannot connect to the miner API - the mining program fails to start. What to do next?
  troubleshoot • • melt  

  1
  0
  Votes
  1
  Posts
  4164
  Views

  

  The miner could not start for any number of reasons. Basically, a read through of the helpme (PiMP Troubleshooter) can tell you whether your GPUs and software is ready to mine. After that, look at what the miner says to see if you have an error. Usually this error is due to user putting wrong information in the conf or pcfg file. To rule that out, we ask you to restore the default that we know works. Run pimpup to be sure you are on the latest pimp software. Run helpme and look at it. Are the GPUs detected properly by all the subsystems? Del your config file. (example: del claymore.dual.pcfg) This will show you if the known-good safe benchmark default configs work. What does pimp --test 1 do? Do you see the miner mining or any error? What does pimp --debug 1 do? Do you see the miner mining or any error? Try adding and running another PiMP managed mining profile and see what that does. Submit your helpme to the forum and/or discord to get some help.
• 

  (Archived) Previous Top Known bugs archive
  • melt  

  1
  0
  Votes
  1
  Posts
  3235
  Views

  

  Overview This list will contain previous support trends listed on Top Known Bugs page, and their fixes. For current Known Bugs: https://forum.getpimp.org/topic/57/top-known-bugs PiMP Does not support 5+ GPU on MSI Z170 boards [Cannot fix] Summary of the bug: Will not boot to PIMP at all. Who it effects: MSI motherboards with 5+ GPU. Status of fix: MSI upstream hardware issue, posted on forum. Workaround: none Network IP not detected on first or second boot on some networks. [Solved] Summary of the bug: First or second boot of PIMP on some networks does not show the IP or allow key entry. Who it effects: AMDGPU users on some networks. Status of fix: Fixed in 2.4 images. Workaround: Find the IP another way, such as looking in the router DHCP table. and SSH the key into the box. ETH DAG hashrate loss on Polaris (RX 4xx series) GPUs [Solved] Summary of the bug: Lower hashrate or hashrate falls after some time mining on AMDGPU. Who it effects: AMDGPU mining Ethereum. Status of fix: Fixed in 2.3.30-amd. Workaround: n/a Clocks/Power not shown on Claymore on 2.3.30-nvidia [Solved] Summary of the bug: 2.3.30-nvidia comes with claymore-dual-v9.7 Who it effects: People using Claymore Dual on Nvidia 2.3.30 Status of fix: Fixed by running pimpup on Claymore dual miner to v10. Workaround: n/a Cannot pimpup due to /tmp/gputype does not exist [Solved] Summary of the bug: There was a bug in a pimpup for a few hours on 20171106 which left this file out. Who it effects: Users who pimpuped imediately on 20171106 Status of fix: Was fixed in pimpup that day. Workaround: Do: echo NVI > /tmp/gputype for Nvidia or echo AMD > /tmp/gputype for AMD and then pimpup --force to force update. AMD Hashrates jumping or amdmeminfo 100% cpu load Summary of the bug: There was a bug in the upstream library amdmeminfo and they have patched it, now you can grab it with the next pimpup. Who it effects: AMD miners. Status of fix: Released in pimpup 20171111 Workaround: Do: pimpup
• 

  Networking Alternatives (Wireless, Static IP, etc)
  • CryptoNews  

  7
  1
  Votes
  7
  Posts
  7988
  Views

  

  @odolea said in Networking Alternatives (Wireless, Static IP, etc): Where can I find the steps to configure wireless connectivity? Hi odolea, thanks for your post. Wireless connection is not recommended for mining because it commonly results in a loss of hashrate/performance. It is especially difficult to validate a pimpkey using wireless. That being said, here are instructions for customizing your network settings: https://forum.getpimp.org/topic/4811/how-to-configure-network-on-pimp-2-8-x-and-above-using-netplan
• 

  Rescue shell
  • CryptoNews  

  1
  0
  Votes
  1
  Posts
  3774
  Views

  

  Overview This only applies to rigs that are not verified with key. PiMP requires networking in order to validate the PiMP key. If PiMP cannot get a network connection in the unverified mode, you will be given what is called the Rescue Shell. This is a special environment designed to do one thing: get you on the network so you can verify. UPDATE! Hot new video how to edit interfaces file: https://youtu.be/Uze50oEU33o Default networking behavior is DHCP PiMP will pull an address from the router using DHCP by default. It is recommended to have a wired ethernet connection for the best mining performance. Wireless support exists, you will have to read up on how to configure it. Also, we have added some example content in the /etc/network/interfaces file to help those users who would like to set their own static IP, etc. What tools are offered in Rescue Shell? First, it will allow you to edit /etc/network/inferfaces directly. Beyond that, some other work may be needed to get networking up. You are given a printout of all pertinent information about the network and network hardware, and a shell to go in and fix it. (Wireless users: this is where iw and iwconfig come into play.) Cannot get a network IP on the first boot? First step is to verify that your network is actually working correctly. Do not assume it is working - you must test it. Disconnect the Ethernet cable from your PiMP machine and connect it directly into a laptop or another computer and verify that it is working. Doing this will verify that your modem, router/gateway, any switches, and all cables are working. With your network verified good, now we can move on to your PiMP machine. PiMP OS uses DHCP as the default for network addressing. Some motherboards take longer than others to initialize their onboard NICs. This "delay" can cause PiMP to timeout and present you with the above image. To get past this, we will need to setup Static network addressing. (Note: Once you have PiMP up and running, you can change back to DHCP if you want.) As seen in the above image, if PiMP OS is not able to find a valid IP, you will be prompted: "Would you like to manually edit your interfaces file? (y/N)". Follow the steps below to establish a Static network address. Answer y followed by <Enter> (PiMP OS will save a backup of the interfaces file and open it for editing. First, we want to disable (adding # to the beginning of a line) of the two (2) DHCP lines. Next, we want to enable (removing the # from the beginning of a line) the eight (8) STATIC lines. Finally, we want to make changes to this data to match our network config. (Note in this example, I've set the Static IP to 192.168.1.209) (You can identify your network's config by inspecting the network addressing of another device on your network.) When complete, your interfaces file should look something close to this: Press Control+X, Y then <Enter> to save the changes. Once back at the command line, type reboot (There is no rb command in Rescue Shell) to reboot your PiMP machine. With a Static IP assigned, you should be presented with the following: Notice that the displayed detected IP address matches what was set in the interfaces file from step 4 above Once your PiMP Key has been validated and your machine reboots to the PiMP OS desktop, feel free to re-edit the /etc/network/interfaces file to re-enable DHCP and to disable STATIC then reboot your PiMP machine.
• 

  amdgpu-pro-pin ERROR: This package can only be installed on Ubuntu 20.04.
  • melt  

  1
  0
  Votes
  1
  Posts
  212
  Views

  

  Hi all, AMD released their 20.04 drivers into their 18.04 repos, sadly this is a breaking change they made which may impact your rigs. If you get any error related to the amdgpu-pro-pin package, like the picture below, here is the fix: apt remove amdgpu-pro-pin Before fix: After fix: If you have any problems with your PiMP software, miners, or Ubuntu updates, let us know so we can guide you! Thanks as always. ~ PiMP DevOps team
• J

  Failed to Load NVIDIA Persistence Daemon
  • jcarpenter  

  4
  0
  Votes
  4
  Posts
  494
  Views

  D

  i ended up re-configuring the card order on the motherboard with NVIDIA cards first and AMD last and everything works
• T

  RDP access
  • thboyes70  

  2
  0
  Votes
  2
  Posts
  297
  Views

  T

  Got this resolved by powering the rig off and back on. Thanks.
• S

  Boots into black screen but can ssh in
  • stephen_chan  

  3
  0
  Votes
  3
  Posts
  868
  Views

  S

  I'm still stuck. I have three GPUs cards 1660 Super 3070 and a 3080 When the rig has the 1660 and 3070, I can boot up, SSH in, and mine: Portable Instant Mining Platform System Info: CPU: Intel(R) Core(TM) i5-7500T CPU @ 2.70GHz [4 Cores @ 800.057 Mhz] Memory: 16 GB IP: 192.168.1.64 WARNING: Mining is not currently supported on Intel GPUS Initialized xorg.conf for nVidia GPUs with BOOST GPUs Detected: GPU 00: INT: Intel Corporation HD Graphics 630 (rev 04): GPU 01: NVI: NVIDIA Corporation Device 2484 (rev a1): GPU-361b72a6-85c0-f529-819f-661527813a2b GPU 09: NVI: NVIDIA Corporation Device 21c4 (rev a1): GPU-b2bc55e4-fc93-a6aa-48b4-e370bb881d0f WARNING: Detected GPU count is 3 - 2 LESS than detected on previous boot! root@192.168.1.64's password: .__ ______ || _____ ______ Portable Instant Mining Platform __ | |/ \ ____ \ PiMP v2.25.2 | |> / | Y Y \ |> / Driver Type: 2.25.2 | /|||_| / / Support: type 'pimp --livechat' || /|| Web: https://getPiMP.org But When I add the 3080, the system doesn't always let me SSH in, but when I can it gives the following: Portable Instant Mining Platform System Info: CPU: Intel(R) Core(TM) i5-7500T CPU @ 2.70GHz [4 Cores @ 800.170 Mhz] Memory: 16 GB IP: 192.168.1.64 Initialized xorg.conf for nVidia GPUs with BOOST GPUs Detected: GPU 01: NVI: NVIDIA Corporation Device 2484 (rev a1): GPU-361b72a6-85c0-f529-819f-661527813a2b GPU 02: NVI: NVIDIA Corporation Device 21c4 (rev a1): GPU-204630c1-64e0-fd21-11c0-4dfd6f79f23c GPU 04: NVI: NVIDIA Corporation Device 21c4 (rev a1): GPU-9fe6f249-2c4a-aea2-9481-55960eaaa252 GPU 05: NVI: GeForce GTX 1060 6GB: GPU-4b268b5c-6e54-05c0-bb23-1e08deefa577 GPU 09: NVI: NVIDIA Corporation Device 21c4 (rev a1): GPU-b2bc55e4-fc93-a6aa-48b4-e370bb881d0f Notice: Detected GPU count is 5 - 3 NEW GPU(s) detected since last boot root@192.168.1.64's password: .__ ______ || _____ ______ Portable Instant Mining Platform __ | |/ \ ____ \ PiMP v2.25.2 | |> / | Y Y \ |> / Driver Type: 2.25.2 | /|||_| / / Support: type 'pimp --livechat' || /|| Web: https://getPiMP.org Thank you very much for supporting getPiMP.org! From the GPU list its clearly found more cards, and I can't figure out what's going on. It can't mine and locks. When I removed the 3080 it mines fine. I thought it was a faulty 3080 and had it replaced but to no-avail. Any ideas?