This has been plaguing me for a couple months but since I used to only hibernate couple times a month I ignored it an hoped it would go away. I used to think it was the DE but after a little googling and discovering pm_trace I now know it's the kernel itself. sync && echo 1 > /sys/power/pm_trace && echo "disk" > /sys/power/state # First run [Sun Jan 13 18:41:34 2019] PM: Basic memory bitmaps created [Sun Jan 13 18:41:34 2019] PM: Preallocating image memory... done (allocated 814796 pages) [Sun Jan 13 18:41:37 2019] PM: Allocated 3259184 kbytes in 2.88 seconds (1131.66 MB/s) [Sun Jan 13 18:41:37 2019] Freezing remaining freezable tasks ... (elapsed 0.096 seconds) done. ... [Sun Jan 13 18:41:52 2019] e1000e: eth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: Rx/Tx [Sun Jan 13 18:42:05 2019] PM: Image saving progress: 10% [Sun Jan 13 18:42:30 2019] PM: Image saving progress: 20% [Sun Jan 13 18:42:51 2019] PM: Image saving progress: 30% [Sun Jan 13 18:43:12 2019] PM: Image saving progress: 40% [Sun Jan 13 18:43:36 2019] PM: Image saving progress: 50% [Sun Jan 13 18:44:02 2019] PM: Image saving progress: 60% [Sun Jan 13 18:44:34 2019] PM: Image saving progress: 70% [Sun Jan 13 18:45:02 2019] PM: Image saving progress: 80% [Sun Jan 13 18:45:45 2019] PM: Image saving progress: 90% [Sun Jan 13 18:46:02 2019] PM: Image saving done [Sun Jan 13 18:46:02 2019] PM: Wrote 3612440 kbytes in 262.50 seconds (13.76 MB/s) [Sun Jan 13 18:46:02 2019] PM: S| # Second run [Sun Jan 13 19:00:21 2019] PM: Basic memory bitmaps created [Sun Jan 13 19:00:21 2019] PM: Preallocating image memory... done (allocated 816831 pages) [Sun Jan 13 19:00:29 2019] PM: Allocated 3267324 kbytes in 7.25 seconds (450.66 MB/s) [Sun Jan 13 19:00:29 2019] Freezing remaining freezable tasks ... (elapsed 0.001 seconds) done. ... [Sun Jan 13 19:00:52 2019] PM: Image saving progress: 10% [Sun Jan 13 19:00:58 2019] e1000e: eth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: Rx/Tx [Sun Jan 13 19:01:12 2019] PM: Image saving progress: 20% [Sun Jan 13 19:01:32 2019] PM: Image saving progress: 30% [Sun Jan 13 19:01:54 2019] PM: Image saving progress: 40% [Sun Jan 13 19:02:21 2019] PM: Image saving progress: 50% [Sun Jan 13 19:02:54 2019] PM: Image saving progress: 60% [Sun Jan 13 19:03:18 2019] PM: Image saving progress: 70% [Sun Jan 13 19:03:46 2019] PM: Image saving progress: 80% [Sun Jan 13 19:04:20 2019] PM: Image saving progress: 90% [Sun Jan 13 19:04:36 2019] PM: Image saving progress: 100% [Sun Jan 13 19:04:36 2019] PM: Image saving done [Sun Jan 13 19:04:36 2019] PM: Wrote 3422232 kbytes in 242.69 seconds (14.10 MB/s) [Sun Jan 13 19:04:36 2019] PM: S| # 3rd run after swapoff, mkswap, swapon (is fragmentation a thing?) [Sun Jan 13 20:20:33 2019] PM: Basic memory bitmaps created [Sun Jan 13 20:20:33 2019] PM: Preallocating image memory... done (allocated 816108 pages) [Sun Jan 13 20:20:41 2019] PM: Allocated 3264432 kbytes in 7.89 seconds (413.74 MB/s) [Sun Jan 13 20:20:41 2019] Freezing remaining freezable tasks ... (elapsed 0.007 seconds) done. ... [Sun Jan 13 20:21:03 2019] PM: Image saving progress: 10% [Sun Jan 13 20:21:20 2019] PM: Image saving progress: 20% [Sun Jan 13 20:21:46 2019] PM: Image saving progress: 30% [Sun Jan 13 20:22:16 2019] PM: Image saving progress: 40% [Sun Jan 13 20:22:43 2019] PM: Image saving progress: 50% [Sun Jan 13 20:23:10 2019] PM: Image saving progress: 60% [Sun Jan 13 20:23:35 2019] PM: Image saving progress: 70% [Sun Jan 13 20:24:01 2019] PM: Image saving progress: 80% [Sun Jan 13 20:24:30 2019] PM: Image saving progress: 90% [Sun Jan 13 20:24:49 2019] PM: Image saving progress: 100% [Sun Jan 13 20:24:49 2019] PM: Image saving done [Sun Jan 13 20:24:49 2019] PM: Wrote 3545300 kbytes in 244.20 seconds (14.51 MB/s) [Sun Jan 13 20:24:49 2019] PM: S| How can this be this slow? My swap partition is the 2nd partition (after boot). I can dump junk over rsync+ssh at 60-80MiB/s onto my almost full ext4 last partition. I thought the first 1/3 or 1/4 on a spinning rust drive was supposed to be the fastest, at least there I should be able to dump sequential writes over 100MiB/s. Where should I look for the problem? It didn't used to take anywhere this long. Maybe a minute and a half tops.
