You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@iggy.apache.org by "k-bx (via GitHub)" <gi...@apache.org> on 2026/06/18 13:16:08 UTC
[I] IO Pressure Stall is near 100% [iggy]
k-bx opened a new issue, #3507:
URL: https://github.com/apache/iggy/issues/3507
### Question
I've deployed fresh iggy on Proxmox container and seeing near 100% IO pressure
<img width="1131" height="440" alt="Image" src="https://github.com/user-attachments/assets/1d76938a-6ec5-4f6a-a71d-ab8307ec819d" />
Instance is a staging container with zero work done. Is this expected from Iggy's side due to io_uring usage?
### Context
_No response_
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@iggy.apache.org.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] IO Pressure Stall is near 100% [iggy]
Posted by "hubcio (via GitHub)" <gi...@apache.org>.
hubcio commented on issue #3507:
URL: https://github.com/apache/iggy/issues/3507#issuecomment-5205442605
@Frtrillo no need to wait - the fix already shipped in compio-driver 0.12.4 and iggy master picked it up in #3720, so any build from current master has it. the `apache/iggy:edge` docker image is built from master, so it has the fix too. kernel >= 6.15 needed for the flag, older kernels keep the old cosmetic behavior.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@iggy.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] IO Pressure Stall is near 100% [iggy]
Posted by "k-bx (via GitHub)" <gi...@apache.org>.
k-bx commented on issue #3507:
URL: https://github.com/apache/iggy/issues/3507#issuecomment-4743021620
Ok redeploying the iggy from master (it was rev 52b060d96d30f39fe7d47c01966c8291b8c789b2 originally but i'll now make it 864ca16a8) and checking:
proxmox host with iggy master running on staging container:
```
$ uname -r
6.17.2-1-pve
```
```
$ sudo iostat -x 2 3
Linux 6.17.2-1-pve (pve-hz-1) 06/18/2026 _x86_64_ (96 CPU)
avg-cpu: %user %nice %system %iowait %steal %idle
9.03 0.00 5.70 0.68 0.00 84.59
Device r/s rkB/s rrqm/s %rrqm r_await rareq-sz w/s wkB/s wrqm/s %wrqm w_await wareq-sz d/s dkB/s drqm/s %drqm d_await dareq-sz f/s f_await aqu-sz %util
nvme0n1 134.07 6641.76 0.00 0.00 0.14 49.54 1577.87 87814.84 0.00 0.00 0.22 55.65 0.77 397.69 0.00 0.00 0.16 518.40 0.00 0.00 0.36 11.56
nvme1n1 135.55 6721.37 0.00 0.00 0.14 49.58 1621.41 87814.84 0.00 0.00 0.20 54.16 0.77 397.68 0.00 0.00 0.16 518.36 0.00 0.00 0.35 11.16
zd0 0.01 0.37 0.00 0.00 0.12 36.61 0.71 10.47 0.00 0.00 0.12 14.78 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 2.55
zd16 1.97 54.36 0.00 0.00 0.11 27.55 10.61 2410.77 0.00 0.00 0.20 227.23 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.92
avg-cpu: %user %nice %system %iowait %steal %idle
17.50 0.00 9.24 10.43 0.00 62.83
Device r/s rkB/s rrqm/s %rrqm r_await rareq-sz w/s wkB/s wrqm/s %wrqm w_await wareq-sz d/s dkB/s drqm/s %drqm d_await dareq-sz f/s f_await aqu-sz %util
nvme0n1 32.50 1328.00 0.00 0.00 0.17 40.86 5399.00 201150.00 0.00 0.00 0.09 37.26 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.46 20.70
nvme1n1 22.00 704.00 0.00 0.00 0.16 32.00 4484.50 201150.00 0.00 0.00 0.13 44.85 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.56 24.95
zd0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
zd16 0.00 0.00 0.00 0.00 0.00 0.00 30.00 1784.00 0.00 0.00 0.07 59.47 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 7.85
```
```
$ sudo cat /sys/fs/cgroup/lxc/660/io.pressure
some avg10=98.98 avg60=76.98 avg300=26.40 total=742329388045
full avg10=93.70 avg60=71.48 avg300=24.38 total=693881020790
```
in container:
```
$ sudo pidstat -d 2 3
Linux 6.17.2-1-pve (ct-echo-staging) 06/18/2026 _x86_64_ (96 CPU)
02:32:12 PM UID PID kB_rd/s kB_wr/s kB_ccwr/s iodelay Command
02:32:14 PM 0 1 0.00 894.04 0.00 0 systemd
02:32:14 PM 101 162 0.00 5.00 0.00 0 rsyslogd
02:32:14 PM 0 499471 0.00 0.56 0.00 0 vector
02:32:14 PM 110 499598 0.00 59.70 0.00 0 postgres
02:32:14 PM 110 499600 0.00 19.90 0.00 0 postgres
02:32:14 PM 110 499719 0.00 71.64 0.00 0 postgres
02:32:14 PM 110 500249 0.00 31.84 0.00 0 postgres
02:32:14 PM 110 504237 0.00 35.82 0.00 0 postgres
02:32:14 PM 110 512121 0.00 3.98 0.00 0 postgres
02:32:14 PM 110 512682 0.00 3.98 0.00 0 postgres
02:32:14 PM 110 513181 0.00 7.96 0.00 0 postgres
02:32:14 PM 0 513250 0.00 15.92 0.00 0 signal-cli-rest
02:32:14 PM 110 2328297 0.00 7.96 0.00 0 postgres
02:32:14 PM 110 2328298 0.00 0.85 0.00 0 postgres
02:32:14 PM 0 2329749 0.00 141.29 0.00 0 systemd-journal
```
```
$ cat /proc/pressure/io
some avg10=34.55 avg60=33.63 avg300=16.70 total=273320399466
full avg10=26.37 avg60=25.53 avg300=12.81 total=214475503133
```
```
$ sudo systemctl stop iggy 2>/dev/null || sudo pkill -x iggy-server
sleep 15
cat /proc/pressure/io
some avg10=7.91 avg60=27.41 avg300=18.52 total=273335573184
full avg10=6.24 avg60=21.05 avg300=14.25 total=214486863147
```
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@iggy.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] IO Pressure Stall is near 100% [iggy]
Posted by "k-bx (via GitHub)" <gi...@apache.org>.
k-bx commented on issue #3507:
URL: https://github.com/apache/iggy/issues/3507#issuecomment-4749206275
Thanks, awaiting for the fix getting upstream and will be happy to test it out. We're evaluating Iggy for non-mission-critical parts of the systems but they run on mission critical servers so polluting the metrics even with fake load numbers is a no go. Using the polling method for now.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@iggy.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] IO Pressure Stall is near 100% [iggy]
Posted by "hubcio (via GitHub)" <gi...@apache.org>.
hubcio commented on issue #3507:
URL: https://github.com/apache/iggy/issues/3507#issuecomment-5507939064
just to add: i'll close this issue once we release 0.9.0 later this week. @k-bx
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@iggy.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] IO Pressure Stall is near 100% [iggy]
Posted by "hubcio (via GitHub)" <gi...@apache.org>.
hubcio commented on issue #3507:
URL: https://github.com/apache/iggy/issues/3507#issuecomment-4749761186
@k-bx you can use that somewhere in your `Cargo.toml` to check if the fix works as expected:
```toml
[patch.crates-io]
compio = { git = "https://github.com/hubcio/compio", rev = "55182653d09d850ec4efe29ab8d9a125442903a0" }
compio-buf = { git = "https://github.com/hubcio/compio", rev = "55182653d09d850ec4efe29ab8d9a125442903a0" }
compio-driver = { git = "https://github.com/hubcio/compio", rev = "55182653d09d850ec4efe29ab8d9a125442903a0" }
compio-executor = { git = "https://github.com/hubcio/compio", rev = "55182653d09d850ec4efe29ab8d9a125442903a0" }
compio-fs = { git = "https://github.com/hubcio/compio", rev = "55182653d09d850ec4efe29ab8d9a125442903a0" }
compio-io = { git = "https://github.com/hubcio/compio", rev = "55182653d09d850ec4efe29ab8d9a125442903a0" }
compio-log = { git = "https://github.com/hubcio/compio", rev = "55182653d09d850ec4efe29ab8d9a125442903a0" }
compio-macros = { git = "https://github.com/hubcio/compio", rev = "55182653d09d850ec4efe29ab8d9a125442903a0" }
compio-net = { git = "https://github.com/hubcio/compio", rev = "55182653d09d850ec4efe29ab8d9a125442903a0" }
compio-quic = { git = "https://github.com/hubcio/compio", rev = "55182653d09d850ec4efe29ab8d9a125442903a0" }
compio-runtime = { git = "https://github.com/hubcio/compio", rev = "55182653d09d850ec4efe29ab8d9a125442903a0" }
compio-tls = { git = "https://github.com/hubcio/compio", rev = "55182653d09d850ec4efe29ab8d9a125442903a0" }
compio-ws = { git = "https://github.com/hubcio/compio", rev = "55182653d09d850ec4efe29ab8d9a125442903a0" }
```
it would be great if you could confirm that the fix works.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@iggy.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] IO Pressure Stall is near 100% [iggy]
Posted by "hubcio (via GitHub)" <gi...@apache.org>.
hubcio commented on issue #3507:
URL: https://github.com/apache/iggy/issues/3507#issuecomment-4742833975
i have a theory, but in order to confirm it i'll need you to provide output of these commands.
goal: show the disk is idle while psi io reads ~100% (io_uring iowait artifact, not real load).
call this on the proxmox host (and save output):
```bash
# kernel: < 6.15 => io_uring idle-wait miscounted as iowait (expected)
uname -r
# disk actually busy? %util ~0 = idle. diskstats is lxcfs-massaged inside the ct, so trust the host
iostat -x 2 3
# the number the proxmox CT graph draws. swap <CTID> for your container id
cat /sys/fs/cgroup/lxc/<CTID>/io.pressure
```
call this inside container (and save output):
```bash
# per-process disk io. iggy-server kB_wr/s ~0 while psi screams = artifact
pidstat -d 2 3
# what lxcfs shows inside (may be host-passthrough on old lxcfs = itself a clue)
cat /proc/pressure/io
# then stop iggy, wait, re-read. collapses => iggy's io_uring waits were the source
systemctl stop iggy 2>/dev/null || pkill -x iggy-server
sleep 15
cat /proc/pressure/io
```
expected if artifact: old-ish kernel, psi io `some` ~100, but `iostat %util` ~0 and `pidstat` iggy ~0. if disk asleep, meter is lying.
please provide output of all these commands.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@iggy.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] IO Pressure Stall is near 100% [iggy]
Posted by "hubcio (via GitHub)" <gi...@apache.org>.
hubcio commented on issue #3507:
URL: https://github.com/apache/iggy/issues/3507#issuecomment-5507832849
@Tchoupinax iggy works fine on 6.12, the floor is 5.19. edge doing nothing for you is expected: the fix keys off `IORING_FEAT_NO_IOWAIT`, which only exists in kernels >= 6.15, so 0.9.0 alone won't change anything on 6.12 either.
as written above, the PSI reading is accounting, not disk load: idle shards (threads running iggy abstractions) parked in `io_uring_enter` are counted as iowait. one addition for your kernel specifically: on 6.12 the menu cpuidle governor still reads iowaiters as a hint to skip deep C-states on those cores, so idle power goes up a bit (6.15 dropped that too). and a graph pegged at 100% hides real io stalls. so a kernel upgrade is worth it for more than the graph.
debian trixie-backports carries a new enough kernel officially (7.1 right now). according to claude:
```bash
echo 'deb http://deb.debian.org/debian trixie-backports main' | sudo tee /etc/apt/sources.list.d/backports.list
sudo apt update
sudo apt install -t trixie-backports linux-image-amd64
sudo reboot
```
should do the trick, but please check it on your side. if iggy runs in a container, the host kernel is the one that counts. no rebuild needed after the reboot, the edge binary detects the feature bit at runtime.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@iggy.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] IO Pressure Stall is near 100% [iggy]
Posted by "Tchoupinax (via GitHub)" <gi...@apache.org>.
Tchoupinax commented on issue #3507:
URL: https://github.com/apache/iggy/issues/3507#issuecomment-5506051726
I'm under Debian 13(.4 or .5) and my kernel version is ~6.12
```
release="6.12.105+deb13-amd64", sysname="Linux", version="#1 SMP PREEMPT_DYNAMIC Debian 6.12.105-1 (2026-08-24)"}
```
Does it mean Iggy recent versions won't work with kernel version lower than 6.15? I'm not even sure you can install it with Debian without forcing package
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@iggy.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] IO Pressure Stall is near 100% [iggy]
Posted by "k-bx (via GitHub)" <gi...@apache.org>.
k-bx commented on issue #3507:
URL: https://github.com/apache/iggy/issues/3507#issuecomment-4743025070
switching from my branch (polling) to upstream
<img width="1163" height="442" alt="Image" src="https://github.com/user-attachments/assets/2af123ac-2840-451c-b4d6-08a7649bc026" />
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@iggy.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] IO Pressure Stall is near 100% [iggy]
Posted by "hubcio (via GitHub)" <gi...@apache.org>.
hubcio commented on issue #3507:
URL: https://github.com/apache/iggy/issues/3507#issuecomment-4748894829
FYI i managed to fix it on my local compio branch. i'll test it further and propose the change for compio devs.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@iggy.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] IO Pressure Stall is near 100% [iggy]
Posted by "Frtrillo (via GitHub)" <gi...@apache.org>.
Frtrillo commented on issue #3507:
URL: https://github.com/apache/iggy/issues/3507#issuecomment-5191886635
got same issue
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@iggy.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] IO Pressure Stall is near 100% [iggy]
Posted by "hubcio (via GitHub)" <gi...@apache.org>.
hubcio commented on issue #3507:
URL: https://github.com/apache/iggy/issues/3507#issuecomment-6032924881
good. im gonna close this issue since problem should be fixed in 0.9.0 release. please open a new one in case something similar occurs :)
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@iggy.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] IO Pressure Stall is near 100% [iggy]
Posted by "hubcio (via GitHub)" <gi...@apache.org>.
hubcio commented on issue #3507:
URL: https://github.com/apache/iggy/issues/3507#issuecomment-4743349143
i think i confirmed it: it's the io_uring iowait accounting artifact, not real disk load.
stopping iggy dropped io PSI `some` 34.55 -> 7.91 (`full` 26 -> 6), yet `iggy-server` never shows up in `pidstat -d` (no disk writes). so iggy drives most of the pressure while writing nothing.
i think it's because each shard (thread) parks in `io_uring_enter` with a poll op armed; kernels >=6.5 charge that wait as iowait, and PSI `io` is derived from it. 6.15 added an opt-out but it's off by default and compio doesn't use it yet - so even a 6.17 kernel still shows it.
notes:
- the container isn't idle (postgres/journald/vector write too) - the leftover `some=7.91` after stopping iggy is their real pressure.
- host `iostat` ~200 MB/s is host-wide (zfs mirror + all guests), not iggy.
is it really an issue? mostly cosmetic - no real disk contention, no harm to iggy or co-tenants. but two real-ish downsides:
- observability gets poisoned: io PSI / iowait are pinned on iggy boxes, so they're useless as a health signal and they make people think iggy is disk-bound (exactly what happened here).
- idle power / freq: false iowait can trip schedutil iowait-boost and historically biased cpuidle toward shallow C-states. probably small on modern kernels, i haven't measured it.
on the PR (#3508): the poll-driver switch does make the graph honest (epoll doesn't set `in_iowait`), so it works as a workaround. but the poll driver runs file writes on a blocking thread pool instead of async io_uring completion, so it trades a cosmetic graph for a real throughput/latency hit under load. for something that doesn't actually harm anything, that's not a trade i want to ship - so i'm going to pass on the poll driver. thanks though, the repro it came with is what pinned this down.
the real fix is upstream in compio: pass `IORING_ENTER_NO_IOWAIT` on the wait (kernels >=6.15 support it). that kills the artifact while keeping io_uring + the perf. i'll investigate this and let you know (prob during the weekend). or, if you want, you can push this further.
until it lands, this is a known-cosmetic thing on iggy boxes - watch real disk metrics (`iostat` / `pidstat`), not psi io.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@iggy.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] IO Pressure Stall is near 100% [iggy]
Posted by "hubcio (via GitHub)" <gi...@apache.org>.
hubcio commented on issue #3507:
URL: https://github.com/apache/iggy/issues/3507#issuecomment-4749651260
raised a PR to compio repository: https://github.com/compio-rs/compio/pull/957 - let's see if they'll allow the change to be merged.i tested it on iggy regression (patched compio on my local machine) and it works just fine. didn't check performance impact. there shouldn't be any.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@iggy.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] IO Pressure Stall is near 100% [iggy]
Posted by "Tchoupinax (via GitHub)" <gi...@apache.org>.
Tchoupinax commented on issue #3507:
URL: https://github.com/apache/iggy/issues/3507#issuecomment-6032391903
@hubcio sorry I thought I had answered. I upgraded kernel as you suggested and it works well now! Thanks you :)
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@iggy.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] IO Pressure Stall is near 100% [iggy]
Posted by "hubcio (via GitHub)" <gi...@apache.org>.
hubcio commented on issue #3507:
URL: https://github.com/apache/iggy/issues/3507#issuecomment-4742442035
hello,
what version of iggy did you deploy?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@iggy.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] IO Pressure Stall is near 100% [iggy]
Posted by "k-bx (via GitHub)" <gi...@apache.org>.
k-bx commented on issue #3507:
URL: https://github.com/apache/iggy/issues/3507#issuecomment-4786395177
@hubcio yeah, looks good
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@iggy.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] IO Pressure Stall is near 100% [iggy]
Posted by "hubcio (via GitHub)" <gi...@apache.org>.
hubcio commented on issue #3507:
URL: https://github.com/apache/iggy/issues/3507#issuecomment-4792363462
nice. so, we wait till `io-uring` is released and then we wait for `compio`.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@iggy.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] IO Pressure Stall is near 100% [iggy]
Posted by "hubcio (via GitHub)" <gi...@apache.org>.
hubcio commented on issue #3507:
URL: https://github.com/apache/iggy/issues/3507#issuecomment-5500620745
@Tchoupinax what's the kernel you ran this on?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@iggy.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] IO Pressure Stall is near 100% [iggy]
Posted by "hubcio (via GitHub)" <gi...@apache.org>.
hubcio commented on issue #3507:
URL: https://github.com/apache/iggy/issues/3507#issuecomment-5500574987
@Tchoupinax it'll be fixed in 0.9.0 release, prob. 5-7 days away from now.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@iggy.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] IO Pressure Stall is near 100% [iggy]
Posted by "Tchoupinax (via GitHub)" <gi...@apache.org>.
Tchoupinax commented on issue #3507:
URL: https://github.com/apache/iggy/issues/3507#issuecomment-5462506554
Hey!
I meet the exact same problem, with an iggy that have absolutely **no** traffic. I tested `apache/iggy:edge` and no amelioration appeared.
<img width="2470" height="805" alt="Image" src="https://github.com/user-attachments/assets/a9bcb847-9580-4cb4-baf4-832f3202afd0" />
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@iggy.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] IO Pressure Stall is near 100% [iggy]
Posted by "hubcio (via GitHub)" <gi...@apache.org>.
hubcio closed issue #3507: IO Pressure Stall is near 100%
URL: https://github.com/apache/iggy/issues/3507
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@iggy.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org