I have a very old (I think circa 2009) Mac Mini, which effectively just acts as a NAS for our TimeMachine backups these days. It was working fine until about 2 days ago, when all of a sudden those backups stopped working. I also couldn't attach to it via Share Screen. When I tried
ssh, I realized that it was running very slowly. It takes 3-5 minutes to
ssh in. Mostly, those time out, but if I can actually get a connection to open, it stays open ... it's just very slow. Even figuring out that I was running El Capitain took almost 6 minutes:
Krotos:~ cwr$ time uname -a Darwin Krotos.local 15.6.0 Darwin Kernel Version 15.6.0: Thu Jun 21 20:07:40 PDT 2018; root:xnu-3248.73.11~1/RELEASE_X86_64 x86_64 real 5m58.602s user 0m0.007s sys 0m0.007s
top takes easily 1-2 minutes before output starts (but then output seems to update at the normal pace).
I assume it's doing something, but ... I can't tell what.
top, when it gets up and running, looks something like this:
Processes: 90 total, 2 running, 17 stuck, 71 sleeping, 365 threads 09:49:05 Load Avg: 1.07, 1.10, 1.07 CPU usage: 0.97% user, 0.48% sys, 98.54% idle SharedLibs: 100M resident, 6980K data, 45M linkedit. MemRegions: 30832 total, 775M resident, 53M private, 153M shared. PhysMem: 2098M used (1007M wired), 6350M unused. VM: 286G vsize, 535M framework vsize, 0(0) swapins, 0(0) swapouts. Networks: packets: 8361/986K in, 1770/314K out. Disks: 11164/837M read, 2740/43M written. PID COMMAND %CPU TIME #TH #WQ #POR MEM PURG CMPR PGRP PPID 281 AppleFileSer 0.0 00:00.20 8 0 78 13M 8192B 0B 281 1 280 top 1.2 00:02.98 1/1 0 22 3044K 0B 0B 280 222 278- fpsaud 0.0 00:00.13 1 0 2 80K 0B 0B 278 1 270 ocspd 0.0 00:00.22 2 0 30 7376K 0B 0B 270 1 269 makewhatis 0.0 00:00.03 1 0 12 584K 0B 0B 245 267 267 sh 0.0 00:00.01 1 0 17 652K 0B 0B 245 2 ...
which, to me, looks like all is well with the world.
Whatever is happening, it appears to be above the network layer, because, for example, I ran continuous
ping while trying to
which grep. I stopped after 100 packets, but it was both fast and didn't drop any:
64 bytes from 192.168.1.10: icmp_seq=101 ttl=64 time=1.315 ms ^C --- krotos.local ping statistics --- 102 packets transmitted, 102 packets received, 0.0% packet loss round-trip min/avg/max/stddev = 1.096/1.927/3.488/0.472 ms
which command, on the other hand ...
Krotos:~ cwr$ time which grep /usr/bin/grep real 1m10.675s user 0m0.007s sys 0m0.005s
Disk space also seems fine, though it takes forever to figure that out:
Krotos:~ cwr$ time df Filesystem 512-blocks Used Available Capacity iused ifree %iused Mounted on /dev/disk0s2 232762432 119458488 112791944 52% 14996309 14098993 52% / devfs 383 383 0 100% 664 0 100% /dev map -hosts 0 0 0 100% 0 0 100% /net map auto_home 0 0 0 100% 0 0 100% /home real 3m4.875s user 0m0.007s sys 0m0.007s
I suppose I could go borrow a monitor from someone and plug it in, but, I'd really like to be able to diagnose this from the command line. I've tried rebooting, and the problem persists. Maybe it's just doing one of those mysterious MacOS things, like re-indexing the spotlight database or something, and it'll be fine in a few days when it finishes doing whatever it's doing, but ... I'd like to know what (and whether that's true and I should just wait a few days, or if it's some bigger problem that won't just go away).
What else can I look at to figure out why simple commands take multiple minutes when the CPU seems to be almost entirely idle and there's plenty of memory and plenty of disk space?