Login

natzilla · 2023-06-26, 06:32 PM

(2023-06-26, 04:35 PM)joshuaboniface Wrote:
(2023-06-26, 03:19 PM)natzilla Wrote:
(2023-06-24, 07:50 AM)joshuaboniface Wrote: OK there's definitely something wrong there, it should not be consuming that much RAM. It's definitely weird that ps aux is reporting so much less than systemctl as well. Unless someone beats me to it I'll check out the logs soon.

I want to add it's possible it might have something to do with transcode, but I'm unsure. Least from what behavior I saw over the weekend.

It definitely seems related though nothing jumped out in the logs either. I'd be really curious what the ps aux and free -m (and htop for good measure) outputs look like while doing transcodes.

Logs, and htop attached during a transcode event

If I'm looking at this right it's using to much cache?

TheDreadPirate · 2023-06-26, 06:38 PM

Would ffmpeg cache if the disk can't write fast enough to keep up with the transcode?

**joshuaboniface** · 2023-06-26, 07:05 PM

Linux caches as much random stuff from the filesystem to RAM as it can, that's what the yellow represents. But that wouldn't be reflected in the output of systemctl. That said that screenshot doesn't look like a transcoding event: I'd expect to see a lot more CPU usage for a transcode (even on a GPU etc.)

natzilla · 2023-06-26, 07:24 PM

(2023-06-26, 07:05 PM)joshuaboniface Wrote: Linux caches as much random stuff from the filesystem to RAM as it can, that's what the yellow represents. But that wouldn't be reflected in the output of systemctl. That said that screenshot doesn't look like a transcoding event: I'd expect to see a lot more CPU usage for a transcode (even on a GPU etc.)

I do have throttle transcodes enabled for my GPU I have on it. I also have remote access limit of 5Mbps for data reasons.

natzilla · 2023-06-26, 07:32 PM

(2023-06-26, 07:05 PM)joshuaboniface Wrote: Linux caches as much random stuff from the filesystem to RAM as it can, that's what the yellow represents. But that wouldn't be reflected in the output of systemctl. That said that screenshot doesn't look like a transcoding event: I'd expect to see a lot more CPU usage for a transcode (even on a GPU etc.)

I've attached another batch of the same commands while starting a transcode since it might have caught it when the transcode was not active.

total used free shared buff/cache available
Mem: 15656 2770 485 17 12400 12537
Swap: 4095 2 4093

USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND
root 1 0.0 0.0 166396 9824 ? Ss Jun25 0:07 /sbin/init
root 2 0.0 0.0 0 0 ? S Jun25 0:00 [kthreadd]
root 3 0.0 0.0 0 0 ? I< Jun25 0:00 [rcu_gp]
root 4 0.0 0.0 0 0 ? I< Jun25 0:00 [rcu_par_gp]
root 5 0.0 0.0 0 0 ? I< Jun25 0:00 [slub_flushwq]
root 6 0.0 0.0 0 0 ? I< Jun25 0:00 [netns]
root 8 0.0 0.0 0 0 ? I< Jun25 0:00 [kworker/0:0H-events_highpri]
root 10 0.0 0.0 0 0 ? I< Jun25 0:00 [mm_percpu_wq]
root 11 0.0 0.0 0 0 ? S Jun25 0:00 [rcu_tasks_rude_]
root 12 0.0 0.0 0 0 ? S Jun25 0:00 [rcu_tasks_trace]
root 13 0.0 0.0 0 0 ? S Jun25 0:00 [ksoftirqd/0]
root 14 0.0 0.0 0 0 ? I Jun25 0:30 [rcu_sched]
root 15 0.0 0.0 0 0 ? S Jun25 0:01 [migration/0]
root 16 0.0 0.0 0 0 ? S Jun25 0:00 [idle_inject/0]
root 18 0.0 0.0 0 0 ? S Jun25 0:00 [cpuhp/0]
root 19 0.0 0.0 0 0 ? S Jun25 0:00 [cpuhp/1]
root 20 0.0 0.0 0 0 ? S Jun25 0:00 [idle_inject/1]
root 21 0.0 0.0 0 0 ? S Jun25 0:01 [migration/1]
root 22 0.0 0.0 0 0 ? S Jun25 0:00 [ksoftirqd/1]
root 24 0.0 0.0 0 0 ? I< Jun25 0:00 [kworker/1:0H-events_highpri]
root 25 0.0 0.0 0 0 ? S Jun25 0:00 [cpuhp/2]
root 26 0.0 0.0 0 0 ? S Jun25 0:00 [idle_inject/2]
root 27 0.0 0.0 0 0 ? S Jun25 0:01 [migration/2]
root 28 0.0 0.0 0 0 ? S Jun25 0:00 [ksoftirqd/2]
root 30 0.0 0.0 0 0 ? I< Jun25 0:00 [kworker/2:0H-events_highpri]
root 31 0.0 0.0 0 0 ? S Jun25 0:00 [cpuhp/3]
root 32 0.0 0.0 0 0 ? S Jun25 0:00 [idle_inject/3]
root 33 0.0 0.0 0 0 ? S Jun25 0:01 [migration/3]
root 34 0.0 0.0 0 0 ? S Jun25 1:44 [ksoftirqd/3]
root 36 0.0 0.0 0 0 ? I< Jun25 0:00 [kworker/3:0H-events_highpri]
root 37 0.0 0.0 0 0 ? S Jun25 0:00 [cpuhp/4]
root 38 0.0 0.0 0 0 ? S Jun25 0:00 [idle_inject/4]
root 39 0.0 0.0 0 0 ? S Jun25 0:01 [migration/4]
root 40 0.0 0.0 0 0 ? S Jun25 0:00 [ksoftirqd/4]
root 42 0.0 0.0 0 0 ? I< Jun25 0:00 [kworker/4:0H-kblockd]
root 43 0.0 0.0 0 0 ? S Jun25 0:00 [cpuhp/5]
root 44 0.0 0.0 0 0 ? S Jun25 0:00 [idle_inject/5]
root 45 0.0 0.0 0 0 ? S Jun25 0:01 [migration/5]
root 46 0.0 0.0 0 0 ? S Jun25 0:00 [ksoftirqd/5]
root 48 0.0 0.0 0 0 ? I< Jun25 0:00 [kworker/5:0H-events_highpri]
root 49 0.0 0.0 0 0 ? S Jun25 0:00 [cpuhp/6]
root 50 0.0 0.0 0 0 ? S Jun25 0:00 [idle_inject/6]
root 51 0.0 0.0 0 0 ? S Jun25 0:01 [migration/6]
root 52 0.0 0.0 0 0 ? S Jun25 0:00 [ksoftirqd/6]
root 54 0.0 0.0 0 0 ? I< Jun25 0:00 [kworker/6:0H-events_highpri]
root 55 0.0 0.0 0 0 ? S Jun25 0:00 [cpuhp/7]
root 56 0.0 0.0 0 0 ? S Jun25 0:00 [idle_inject/7]
root 57 0.0 0.0 0 0 ? S Jun25 0:01 [migration/7]
root 58 0.0 0.0 0 0 ? S Jun25 0:02 [ksoftirqd/7]
root 60 0.0 0.0 0 0 ? I< Jun25 0:00 [kworker/7:0H-events_highpri]
root 61 0.0 0.0 0 0 ? S Jun25 0:00 [kdevtmpfs]
root 62 0.0 0.0 0 0 ? I< Jun25 0:00 [inet_frag_wq]
root 63 0.0 0.0 0 0 ? S Jun25 0:00 [kauditd]
root 65 0.0 0.0 0 0 ? S Jun25 0:00 [khungtaskd]
root 66 0.0 0.0 0 0 ? S Jun25 0:00 [oom_reaper]
root 67 0.0 0.0 0 0 ? I< Jun25 0:00 [writeback]
root 68 0.0 0.0 0 0 ? S Jun25 0:13 [kcompactd0]
root 69 0.0 0.0 0 0 ? SN Jun25 0:00 [ksmd]
root 70 0.0 0.0 0 0 ? SN Jun25 0:02 [khugepaged]
root 118 0.0 0.0 0 0 ? I< Jun25 0:00 [kintegrityd]
root 119 0.0 0.0 0 0 ? I< Jun25 0:00 [kblockd]
root 120 0.0 0.0 0 0 ? I< Jun25 0:00 [blkcg_punt_bio]
root 121 0.0 0.0 0 0 ? I< Jun25 0:00 [tpm_dev_wq]
root 122 0.0 0.0 0 0 ? I< Jun25 0:00 [ata_sff]
root 123 0.0 0.0 0 0 ? I< Jun25 0:00 [md]
root 124 0.0 0.0 0 0 ? I< Jun25 0:00 [edac-poller]
root 125 0.0 0.0 0 0 ? I< Jun25 0:00 [devfreq_wq]
root 126 0.0 0.0 0 0 ? S Jun25 0:00 [watchdogd]
root 129 0.0 0.0 0 0 ? I< Jun25 0:01 [kworker/6:1H-kblockd]
root 131 0.0 0.0 0 0 ? S Jun25 0:27 [kswapd0]
root 132 0.0 0.0 0 0 ? S Jun25 0:00 [ecryptfs-kthrea]
root 134 0.0 0.0 0 0 ? I< Jun25 0:00 [kthrotld]
root 135 0.0 0.0 0 0 ? I< Jun25 0:00 [acpi_thermal_pm]
root 137 0.0 0.0 0 0 ? S Jun25 0:00 [scsi_eh_0]
root 138 0.0 0.0 0 0 ? I< Jun25 0:00 [scsi_tmf_0]
root 139 0.0 0.0 0 0 ? S Jun25 0:00 [scsi_eh_1]
root 140 0.0 0.0 0 0 ? I< Jun25 0:00 [scsi_tmf_1]
root 142 0.0 0.0 0 0 ? I< Jun25 0:00 [vfio-irqfd-clea]
root 143 0.0 0.0 0 0 ? I< Jun25 0:00 [mld]
root 144 0.0 0.0 0 0 ? I< Jun25 0:02 [kworker/5:1H-kblockd]
root 145 0.0 0.0 0 0 ? I< Jun25 0:00 [ipv6_addrconf]
root 151 0.0 0.0 0 0 ? I< Jun25 0:02 [kworker/4:1H-kblockd]
root 159 0.0 0.0 0 0 ? I< Jun25 0:00 [kstrp]
root 162 0.0 0.0 0 0 ? I< Jun25 0:00 [zswap-shrink]
root 168 0.0 0.0 0 0 ? I< Jun25 0:00 [charger_manager]
root 191 0.0 0.0 0 0 ? I< Jun25 0:02 [kworker/2:1H-kblockd]
root 221 0.0 0.0 0 0 ? I< Jun25 0:07 [kworker/0:1H-kblockd]
root 223 0.0 0.0 0 0 ? S Jun25 0:00 [scsi_eh_2]
root 224 0.0 0.0 0 0 ? I< Jun25 0:00 [scsi_tmf_2]
root 225 0.0 0.0 0 0 ? I< Jun25 0:00 [cryptd]
root 242 0.0 0.0 0 0 ? I< Jun25 0:03 [kworker/3:1H-kblockd]
root 245 0.0 0.0 0 0 ? I< Jun25 0:02 [kworker/1:1H-kblockd]
root 258 0.0 0.0 0 0 ? I< Jun25 0:02 [kworker/7:1H-kblockd]
root 302 0.0 0.0 0 0 ? I< Jun25 0:00 [raid5wq]
root 350 0.0 0.0 0 0 ? S Jun25 0:38 [jbd2/sda2-8]
root 351 0.0 0.0 0 0 ? I< Jun25 0:00 [ext4-rsv-conver]
root 427 0.0 0.3 137952 52720 ? S<s Jun25 0:43 /lib/systemd/systemd-journald
root 463 0.0 0.0 0 0 ? I< Jun25 0:00 [kaluad]
root 465 0.0 0.0 0 0 ? I< Jun25 0:00 [kmpath_rdacd]
root 466 0.0 0.0 0 0 ? I< Jun25 0:00 [kmpathd]
root 467 0.0 0.0 0 0 ? I< Jun25 0:00 [kmpath_handlerd]
root 471 0.0 0.1 289348 27100 ? SLsl Jun25 0:26 /sbin/multipathd -d -s
root 481 0.0 0.0 26020 5180 ? Ss Jun25 0:01 /lib/systemd/systemd-udevd
_rpc 547 0.0 0.0 8100 4104 ? Ss Jun25 0:00 /sbin/rpcbind -f -w
systemd+ 556 0.0 0.0 89352 5016 ? Ssl Jun25 0:00 /lib/systemd/systemd-timesyncd
root 559 0.0 0.0 0 0 ? I< Jun25 0:00 [rpciod]
root 560 0.0 0.0 0 0 ? I< Jun25 0:00 [xprtiod]
systemd+ 703 0.0 0.0 16116 6544 ? Ss Jun25 0:02 /lib/systemd/systemd-networkd
root 705 0.0 0.0 0 0 ? S Jun25 0:00 [nv_queue]
root 706 0.0 0.0 0 0 ? S Jun25 0:00 [nv_queue]
systemd+ 708 0.0 0.0 25528 11248 ? Ss Jun25 0:00 /lib/systemd/systemd-resolved
root 710 0.0 0.0 0 0 ? S Jun25 0:00 [nvidia-modeset/]
root 711 0.0 0.0 0 0 ? S Jun25 0:00 [nvidia-modeset/]
root 721 0.0 0.0 0 0 ? S Jun25 0:00 [UVM global queu]
root 722 0.0 0.0 0 0 ? S Jun25 0:00 [UVM deferred re]
root 723 0.0 0.0 0 0 ? S Jun25 0:00 [UVM Tools Event]
message+ 745 0.0 0.0 8844 4548 ? Ss Jun25 0:00 @dbus-daemon --system --address=systemd: --nofork --nopidfile --systemd-activation --syslog-only
root 750 0.0 0.0 82772 3880 ? Ssl Jun25 0:12 /usr/sbin/irqbalance --foreground
jellyfin 751 1.9 11.8 7238272 1894220 ? Ssl Jun25 47:55 /usr/bin/jellyfin --webdir=/usr/share/jellyfin/web --restartpath=/usr/lib/jellyfin/restart.sh --ffmpeg=/usr/lib/jellyfin-ffmpeg/ffmpeg
root 752 0.0 0.0 32796 14444 ? Ss Jun25 0:00 /usr/bin/python3 /usr/bin/networkd-dispatcher --run-startup-triggers
nvidia-+ 754 0.0 0.0 5308 1904 ? Ss Jun25 0:00 /usr/bin/nvidia-persistenced --user nvidia-persistenced --no-persistence-mode --verbose
root 755 0.0 0.0 234492 6684 ? Ssl Jun25 0:00 /usr/libexec/polkitd --no-debug
syslog 756 0.0 0.0 222400 5548 ? Ssl Jun25 0:06 /usr/sbin/rsyslogd -n -iNONE
root 757 0.0 0.0 0 0 ? I< Jun25 0:00 [nfsiod]
root 764 0.0 0.2 1393328 40184 ? Ssl Jun25 0:25 /usr/lib/snapd/snapd
root 768 0.0 0.0 48108 5872 ? Ss Jun25 0:00 /lib/systemd/systemd-logind
root 775 0.0 0.0 392576 11556 ? Ssl Jun25 0:00 /usr/libexec/udisks2/udisksd
root 783 0.0 0.0 15420 6408 ? Ss Jun25 0:00 sshd: /usr/sbin/sshd -D [listener] 0 of 10-100 startups
root 797 0.0 0.0 317956 11684 ? Ssl Jun25 0:00 /usr/sbin/ModemManager
root 812 0.0 0.1 109748 17172 ? Ssl Jun25 0:00 /usr/bin/python3 /usr/share/unattended-upgrades/unattended-upgrade-shutdown --wait-for-signal
root 834 0.0 0.0 0 0 ? S Jun25 0:00 [NFSv4 callback]
root 848 1.8 0.4 799044 71956 ? Ssl Jun25 44:27 /usr/sbin/tailscaled --state=/var/lib/tailscale/tailscaled.state --socket=/run/tailscale/tailscaled.sock --port=41641
root 864 0.0 0.0 6892 2664 ? Ss Jun25 0:00 /usr/sbin/cron -f -P
daemon 868 0.0 0.0 3860 1284 ? Ss Jun25 0:00 /usr/sbin/atd -f
root 882 0.0 0.0 6172 996 tty1 Ss+ Jun25 0:00 /sbin/agetty -o -p -- \u --noclear tty1 linux
root 3256 0.0 0.0 239608 7508 ? Ssl Jun25 0:00 /usr/libexec/upowerd
root 6469 0.0 0.0 17168 8264 ? Ss Jun25 0:00 sshd: pfs [priv]
pfs 6472 0.0 0.0 17052 7564 ? Ss Jun25 0:00 /lib/systemd/systemd --user
pfs 6473 0.0 0.0 169448 3860 ? S Jun25 0:00 (sd-pam)
pfs 6556 0.0 0.0 17300 6020 ? R Jun25 0:03 sshd: pfs@pts/0
pfs 6557 0.0 0.0 8732 4964 pts/0 Ss Jun25 0:00 -bash
root 12402 0.0 0.0 0 0 ? I 05:25 0:00 [kworker/2:1-mm_percpu_wq]
root 13530 0.0 0.0 0 0 ? I 07:44 0:00 [kworker/4:1-events]
root 13715 0.0 0.0 0 0 ? I 08:21 0:00 [kworker/6:0-events]
root 13776 0.0 0.0 0 0 ? I 08:33 0:01 [kworker/4:0-events]
root 14086 0.0 0.0 0 0 ? I 09:32 0:00 [kworker/7:2-mm_percpu_wq]
root 14473 0.0 0.0 0 0 ? I 10:47 0:00 [kworker/1:2-events]
root 14932 0.0 0.0 0 0 ? I 11:33 0:00 [kworker/5:0-events]
root 15001 0.0 0.0 0 0 ? I 11:42 0:00 [kworker/6:2-events]
root 15661 0.0 0.0 0 0 ? I 13:07 0:00 [kworker/3:0-events]
root 16845 0.0 0.1 295552 20332 ? Ssl 14:39 0:00 /usr/libexec/packagekitd
root 18396 0.1 0.0 0 0 ? I 17:32 0:10 [kworker/u16:7-rpciod]
root 18451 0.0 0.0 0 0 ? I 17:41 0:00 [kworker/0:2-events]
root 18661 0.2 0.0 0 0 ? I 18:21 0:12 [kworker/u16:4-rpciod]
root 18702 0.0 0.0 0 0 ? I 18:28 0:00 [kworker/3:2-events]
root 18717 0.0 0.0 0 0 ? I 18:31 0:00 [kworker/1:1-ata_sff]
root 18747 0.0 0.0 0 0 ? I 18:36 0:00 [kworker/0:0-events]
root 18807 0.3 0.0 0 0 ? I 18:43 0:08 [kworker/u16:1-rpciod]
root 18808 0.3 0.0 0 0 ? I 18:43 0:10 [kworker/u16:2-events_unbound]
root 18814 0.3 0.0 0 0 ? I 18:43 0:10 [kworker/u16:3-rpciod]
root 18816 0.3 0.0 0 0 ? I 18:43 0:11 [kworker/u16:6-nfsiod]
root 18842 0.0 0.0 0 0 ? I 18:49 0:00 [kworker/7:1-events]
root 18843 0.4 0.0 0 0 ? I 18:49 0:10 [kworker/u16:0-rpciod]
root 18902 0.9 0.0 0 0 ? I< 18:54 0:20 [kworker/u17:0-xprtiod]
root 18932 0.6 0.0 0 0 ? I< 18:54 0:14 [kworker/u17:2-xprtiod]
jellyfin 18936 9.2 2.1 10036084 351332 ? Sl 18:54 3:17 /usr/lib/jellyfin-ffmpeg/ffmpeg -analyzeduration 50000000 -init_hw_device cuda=cu:0 -filter_hw_device cu -hwaccel cuda -hwaccel_output_format cuda -threads 1 -a
root 18937 0.3 0.0 0 0 ? S 18:54 0:08 [irq/38-nvidia]
root 18938 0.0 0.0 0 0 ? S 18:54 0:00 [nvidia]
root 18939 0.1 0.0 0 0 ? S 18:54 0:02 [nv_queue]
root 18940 0.0 0.0 0 0 ? S 18:54 0:00 [UVM GPU1 BH]
root 19052 0.0 0.0 0 0 ? I 19:04 0:00 [kworker/5:1-events]
root 19095 0.4 0.0 0 0 ? I 19:12 0:04 [kworker/u16:5-events_unbound]
root 19120 0.0 0.0 0 0 ? I 19:17 0:00 [kworker/2:0-events]
root 19121 0.4 0.0 0 0 ? I 19:17 0:03 [kworker/u16:8-rpciod]
root 19132 0.6 0.0 0 0 ? I< 19:20 0:03 [kworker/u17:3-xprtiod]
jellyfin 19150 7.8 0.2 233988 37400 ? Sl 19:25 0:26 /usr/lib/jellyfin-ffmpeg/ffmpeg -analyzeduration 50000000 -fflags +genpts -i file Confused-face

mnt/media/Shows/All Shows/That 70's Show/That.70s.Show.S01.Season.1.Complete.
jellyfin 19196 43.7 1.7 9929860 276252 ? Rl 19:30 0:03 /usr/lib/jellyfin-ffmpeg/ffmpeg -analyzeduration 50000000 -init_hw_device cuda=cu:0 -filter_hw_device cu -hwaccel cuda -hwaccel_output_format cuda -threads 1 -a
pfs 19217 0.0 0.0 10068 1568 pts/0 R+ 19:30 0:00 ps aux

**joshuaboniface** · 2023-06-27, 03:31 AM

Yep definitely transcoding there, and RAM usage is still relatively low at only 2.73G (the green portion of the RAM graph). So that looks totally normal.

I think we'd have to wait until you see it fully fill up again and then grab snapshots of as much of those tools as possible to see what it might be.

natzilla · 2023-06-27, 04:33 PM

(2023-06-27, 03:31 AM)joshuaboniface Wrote: Yep definitely transcoding there, and RAM usage is still relatively low at only 2.73G (the green portion of the RAM graph). So that looks totally normal.

I think we'd have to wait until you see it fully fill up again and then grab snapshots of as much of those tools as possible to see what it might be.

Hmm, generally I just get reports from my family users that the responses to requests are extremely delayed, or never does anything when you select content. So I can wait for that to happen again. I did adjust the transcoding cache clean filter to be every 6 hours from 24. So we will see what happens.

Login

Username/Email:
Password:


Or login with a social network below

Login

Username/Email:
Password:


Or login with a social network below

RAM Usage