trohn_javolta Posted April 3, 2018 Posted April 3, 2018 Hi, yesterday I did a apt upgrade and noticed that my omv install also got updated. Now copying from my HC2 I only get like 50 mb/s. I always got like 100 mb/s..at least win explorer showed this, I know it's not accurate. Could the update have erased/overwritten all these nice tweaks @tkaiser made for us to increase tranfer speed? That would be very unfortunate :/ Any chance I can roll back the omv update?
tkaiser Posted April 3, 2018 Posted April 3, 2018 Can you provide 'armbianmonitor -u' output please...
trohn_javolta Posted April 3, 2018 Author Posted April 3, 2018 S 6 minutes ago, tkaiser said: Can you provide 'armbianmonitor -u' output please... Sure: http://ix.io/16d1
tkaiser Posted April 3, 2018 Posted April 3, 2018 16 minutes ago, trohn_javolta said: Sure: http://ix.io/16d1 Thank you! Nothing suspicious except your H2 running quite hot (76.0°C) and some messages talking about syn flooding. I would try to 'gently massage' the PCB to improve contact between SoC and enclosure (when I got my HC2 this helped to decrease temperatures a bit -- obviously a sign of thermal paste not applied that great). Cpufreq scaling works, you still run the old kernel and the OMV updates itself haven't change much. So I would better do some isolated tests (iozone for your storage and iperf3 to test between different hosts for networking anomalies).
trohn_javolta Posted April 3, 2018 Author Posted April 3, 2018 3 minutes ago, tkaiser said: Thank you! Nothing suspicious except your H2 running quite hot (76.0°C) and some messages talking about syn flooding. I would try to 'gently massage' the PCB to improve contact between SoC and enclosure (when I got my HC2 this helped to decrease temperatures a bit -- obviously a sign of thermal paste not applied that great). Cpufreq scaling works, you still run the old kernel and the OMV updates itself haven't change much. So I would better do some isolated tests (iozone for your storage and iperf3 to test between different hosts for networking anomalies). I read that and also tried to massage the soc to improve the contact. Last time I checked, temp was ok but yeah the board recently did some downloading and copying. I'll think about reapplying thermal paste. Can't be that hard difficult right? I'd buy some thermal paste, take the soc of the enclosure, clean the old thermal paste with some rubbing alcohol and apply new none. Old kernel...yeah, I don't know how to switch to the new one or if I even should.. I thought if it's a good idea to change, arbian will do it automatically after apt upgrade. I don't really know how to do these tests with iozon and iperf3, I can monitor it and hope the tranfer speed recovers.
trohn_javolta Posted April 4, 2018 Author Posted April 4, 2018 Hmm... now transfer speed is alright again... Could low transfer speed be related to higher cpu load and temperature? I mean I don't really have cpu intensive stuff running on my HC2 but maybe higher temperature makes it clock down? I'll definetly reapply thermal paste.... it's sad that they don't do that properly ex works.
tkaiser Posted April 4, 2018 Posted April 4, 2018 16 minutes ago, trohn_javolta said: Could low transfer speed be related to higher cpu load and temperature? Partially it could. If all CPU cores are busy doing stuff then adding another task might slow everything down (more tasks than resources). When throttling jumps in transfer speeds definitely decrease but it's unlikely this happened (I mentioned the thermal readouts only since I think they're too high). Again: check your dmesg output for 'SYN flooding'. If there's something in your network hammering the machine then for sure throughput of other services will then being bottlenecked by a single GbE connection. And no, I don't know what you had running on port 9090 or 8888 (the latter most probably being RPi-Monitor?)
trohn_javolta Posted April 14, 2018 Author Posted April 14, 2018 On 4.4.2018 at 11:04 AM, tkaiser said: Partially it could. If all CPU cores are busy doing stuff then adding another task might slow everything down (more tasks than resources). When throttling jumps in transfer speeds definitely decrease but it's unlikely this happened (I mentioned the thermal readouts only since I think they're too high). Again: check your dmesg output for 'SYN flooding'. If there's something in your network hammering the machine then for sure throughput of other services will then being bottlenecked by a single GbE connection. And no, I don't know what you had running on port 9090 or 8888 (the latter most probably being RPi-Monitor?) Got rid of the syn flooding and also reapplied thermal paste. I saw that hardkernel (or better said the factory where the boards are produced) did a good job on the paste. I noticed temp really spikes up when I tranfer sth. via network, from like 55 to over 70 degree. Is that to worry? I don't copy sth. at full speed all the time. I only notice it via smb, via nfs there isn't suach a spike in temp.
Recommended Posts