Silent stoppage building www/chromium
bob prohaska
fbsd at www.zefox.net
Sun Aug 30 15:48:26 UTC 2020
Trying again to compile www/chromium on a Pi3B. Ports are at 546965,
system is at r364900. Root is a USB mechanical hard disk, with swap
on both hard disk and microSD.
Something, not sure what, seems to stall silently. The host is still
responsive, but sluggish, as if something is very busy. Top reports
last pid: 4819; load averages: 2.65, 2.33, 1.78 up 0+14:24:26 08:40:45
52 processes: 2 running, 50 sleeping
CPU: 55.2% user, 0.0% nice, 3.2% system, 1.2% interrupt, 40.4% idle
Mem: 300M Active, 120M Inact, 28M Laundry, 176M Wired, 98M Buf, 280M Free
Swap: 6042M Total, 589M Used, 5453M Free, 9% Inuse, 724K In
PID USERNAME THR PRI NICE SIZE RES STATE C TIME WCPU COMMAND
3264 root 1 21 0 20M 644K select 2 0:00 0.00% sshd
4720 root 1 52 0 13M 0B wait 2 0:00 0.00% <gmake>
928 root 1 52 0 13M 424K ttyin 0 0:00 0.00% getty
97 root 1 52 0 12M 0B pause 0 0:00 0.00% <adjkerntz>
929 root 1 52 0 13M 424K ttyin 2 0:00 0.00% getty
961 bob 1 20 0 13M 0B wait 1 0:00 0.00% <su>
4802 root 5 22 0 245M 130M uwait 3 4:54 99.87% rustc
4818 root 1 23 0 113M 53M wait 0 0:01 0.00% clang-9
722 root 1 20 0 13M 784K select 2 0:01 0.00% syslogd
866 root 1 20 0 20M 572K select 3 0:01 0.00% sshd
930 root 1 52 0 13M 424K ttyin 3 0:00 0.00% getty
962 root 1 20 0 14M 0B pause 3 0:00 0.00% <csh>
3267 bob 1 20 0 20M 824K select 0 0:00 0.00% sshd
4595 bob 1 20 0 13M 0B wait 2 0:00 0.00% <su>
Gstat reports
dT: 1.013s w: 1.000s
L(q) ops/s r/s kBps ms/r w/s kBps ms/w %busy Name
0 44 44 265 1.5 0 0 0.0 6.8| mmcsd0
1 38 38 178 19.2 0 0 0.0 67.7| da0
0 0 0 0 0.0 0 0 0.0 0.0| mmcsd0s1
0 44 44 265 1.6 0 0 0.0 7.2| mmcsd0s2
0 0 0 0 0.0 0 0 0.0 0.0| msdosfs/MSDOSBOOT
0 0 0 0 0.0 0 0 0.0 0.0| mmcsd0s2a
0 44 44 265 1.7 0 0 0.0 7.3| mmcsd0s2b
0 0 0 0 0.0 0 0 0.0 0.0| ufsid/5bbd5a23da8b6195
0 0 0 0 0.0 0 0 0.0 0.0| ufs/rootfs
0 0 0 0 0.0 0 0 0.0 0.0| da0s1
1 38 38 178 19.3 0 0 0.0 67.9| da0s2
0 2 2 12 35.6 0 0 0.0 7.0| da0s2a
1 37 37 166 18.4 0 0 0.0 67.2| da0s2b
Both swap devices are busy, but neither seems saturated and much higher
swap usage has been observed before without the system getting stuck.
If anybody can suggest something to try or explore please do.
Thanks for reading,
bob prohaska
More information about the freebsd-ports
mailing list