Koozali.org: home of the SME Server

perf: interrupt took too long

Offline gieres

  • *
  • 213
  • +0/-0
perf: interrupt took too long
« on: February 05, 2022, 09:54:37 PM »
Bonsoir,
Et de nouveaux messages d'erreurs !
Code: [Select]
Feb  3 22:21:14 kooz1 kernel: [52560.297522] perf: interrupt took too long (2548 > 2500), lowering kernel.perf_event_max_sample_rate to 78000
Feb  4 03:27:08 kooz1 rsyslogd:  [origin software="rsyslogd" swVersion="8.24.0-57.el7_9.1" x-pid="2016" x-info="http://www.rsyslog.com"] rsyslogd was HUPed
Feb  4 11:34:40 kooz1 kernel: [100167.513135] e1000e 0000:00:19.0 eno1: Detected Hardware Unit Hang:
Feb  4 11:34:40 kooz1 kernel: [100167.513135]   TDH                  <14>
Feb  4 11:34:40 kooz1 kernel: [100167.513135]   TDT                  <47>
Feb  4 11:34:40 kooz1 kernel: [100167.513135]   next_to_use          <47>
Feb  4 11:34:40 kooz1 kernel: [100167.513135]   next_to_clean        <12>
Feb  4 11:34:40 kooz1 kernel: [100167.513135] buffer_info[next_to_clean]:
Feb  4 11:34:40 kooz1 kernel: [100167.513135]   time_stamp           <105f3c870>
Feb  4 11:34:40 kooz1 kernel: [100167.513135]   next_to_watch        <14>
Feb  4 11:34:40 kooz1 kernel: [100167.513135]   jiffies              <105f3d080>
Feb  4 11:34:40 kooz1 kernel: [100167.513135]   next_to_watch.status <0>
Feb  4 11:34:40 kooz1 kernel: [100167.513135] MAC Status             <40080083>
Feb  4 11:34:40 kooz1 kernel: [100167.513135] PHY Status             <796d>
Feb  4 11:34:40 kooz1 kernel: [100167.513135] PHY 1000BASE-T Status  <3800>
Feb  4 11:34:40 kooz1 kernel: [100167.513135] PHY Extended Status    <3000>
Feb  4 11:34:40 kooz1 kernel: [100167.513135] PCI Status             <10>
Feb  4 11:34:42 kooz1 kernel: [100169.513177] e1000e 0000:00:19.0 eno1: Detected Hardware Unit Hang:
Feb  4 11:34:42 kooz1 kernel: [100169.513177]   TDH                  <14>
Feb  4 11:34:42 kooz1 kernel: [100169.513177]   TDT                  <47>
Feb  4 11:34:42 kooz1 kernel: [100169.513177]   next_to_use          <47>
Feb  4 11:34:42 kooz1 kernel: [100169.513177]   next_to_clean        <12>
Feb  4 11:34:42 kooz1 kernel: [100169.513177] buffer_info[next_to_clean]:
Feb  4 11:34:42 kooz1 kernel: [100169.513177]   time_stamp           <105f3c870>
Feb  4 11:34:42 kooz1 kernel: [100169.513177]   next_to_watch        <14>
Feb  4 11:34:42 kooz1 kernel: [100169.513177]   jiffies              <105f3d850>
Feb  4 11:34:42 kooz1 kernel: [100169.513177]   next_to_watch.status <0>
Feb  4 11:34:42 kooz1 kernel: [100169.513177] MAC Status             <40080083>
Feb  4 11:34:42 kooz1 kernel: [100169.513177] PHY Status             <796d>
Feb  4 11:34:42 kooz1 kernel: [100169.513177] PHY 1000BASE-T Status  <3800>
Feb  4 11:34:42 kooz1 kernel: [100169.513177] PHY Extended Status    <3000>
Feb  4 11:34:42 kooz1 kernel: [100169.513177] PCI Status             <10>
Feb  4 11:34:44 kooz1 kernel: [100171.512496] ------------[ cut here ]------------
Feb  4 11:34:44 kooz1 kernel: [100171.512518] WARNING: CPU: 1 PID: 0 at net/sched/sch_generic.c:356 dev_watchdog+0x248/0x260
Feb  4 11:34:44 kooz1 kernel: [100171.512523] NETDEV WATCHDOG: eno1 (e1000e): transmit queue 0 timed out
Feb  4 11:34:44 kooz1 kernel: [100171.512525] Modules linked in: fuse btrfs raid6_pq xor vfat msdos fat ipt_REJECT nf_reject_ipv4 xt_recent ipt_MASQUERADE nf_nat_masquerade_ipv4 xt_conntrack xt_nat nf_nat_ftp nf_conntrack_ftp iptable_mangle iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack iptable_filter ipt_ULOG sunrpc ext4 mbcache jbd2 intel_powerclamp coretemp intel_rapl iosf_mbi kvm_intel kvm irqbypass crc32_pclmul crc32c_intel ghash_clmulni_intel iTCO_wdt iTCO_vendor_support cryptd ipmi_ssif joydev pcspkr ipmi_si sg ipmi_devintf i2c_i801 ipmi_msghandler e1000e video ptp pps_core lpc_ich ie31200_edac dummy ip_tables xfs libcrc32c sd_mod crc_t10dif crct10dif_generic mgag200 i2c_algo_bit drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops ttm drm ahci libahci crct10dif_pclmul crct10dif_common libata serio_raw megaraid_sas drm_panel_orientation_quirks dm_mirror dm_region_hash dm_log dm_mod
Feb  4 11:34:44 kooz1 kernel: [100171.512602] CPU: 1 PID: 0 Comm: swapper/1 Not tainted 3.10.0-1160.53.1.el7.x86_64 #1
Feb  4 11:34:44 kooz1 kernel: [100171.512606] Hardware name: Supermicro X9SCL/X9SCM/X9SCL/X9SCM, BIOS 2.10 01/09/2014
Feb  4 11:34:44 kooz1 kernel: [100171.512609] Call Trace:
Feb  4 11:34:44 kooz1 kernel: [100171.512614]  <IRQ>  [<ffffffff99f83579>] dump_stack+0x19/0x1b
Feb  4 11:34:44 kooz1 kernel: [100171.512626]  [<ffffffff9989b278>] __warn+0xd8/0x100
Feb  4 11:34:44 kooz1 kernel: [100171.512633]  [<ffffffff9989b2ff>] warn_slowpath_fmt+0x5f/0x80
Feb  4 11:34:44 kooz1 kernel: [100171.512643]  [<ffffffff99e83f78>] dev_watchdog+0x248/0x260
Feb  4 11:34:44 kooz1 kernel: [100171.512648]  [<ffffffff99e83d30>] ? dev_deactivate_queue.constprop.27+0x60/0x60
Feb  4 11:34:44 kooz1 kernel: [100171.512654]  [<ffffffff998abdb8>] call_timer_fn+0x38/0x110
Feb  4 11:34:44 kooz1 kernel: [100171.512659]  [<ffffffff99e83d30>] ? dev_deactivate_queue.constprop.27+0x60/0x60
Feb  4 11:34:44 kooz1 kernel: [100171.512664]  [<ffffffff998ae3cd>] run_timer_softirq+0x25d/0x340
Feb  4 11:34:44 kooz1 kernel: [100171.512670]  [<ffffffff998a4bf5>] __do_softirq+0xf5/0x280
Feb  4 11:34:44 kooz1 kernel: [100171.512676]  [<ffffffff99f994ec>] call_softirq+0x1c/0x30
Feb  4 11:34:44 kooz1 kernel: [100171.512682]  [<ffffffff9982f715>] do_softirq+0x65/0xa0
Feb  4 11:34:44 kooz1 kernel: [100171.512687]  [<ffffffff998a4f75>] irq_exit+0x105/0x110
Feb  4 11:34:44 kooz1 kernel: [100171.512693]  [<ffffffff99f9aa28>] smp_apic_timer_interrupt+0x48/0x60
Feb  4 11:34:44 kooz1 kernel: [100171.512698]  [<ffffffff99f96fba>] apic_timer_interrupt+0x16a/0x170
Feb  4 11:34:44 kooz1 kernel: [100171.512701]  <EOI>  [<ffffffff99dc8dc7>] ? cpuidle_enter_state+0x57/0xd0
Feb  4 11:34:44 kooz1 kernel: [100171.512711]  [<ffffffff99dc8f1e>] cpuidle_idle_call+0xde/0x230
Feb  4 11:34:44 kooz1 kernel: [100171.512717]  [<ffffffff99837c8e>] arch_cpu_idle+0xe/0xc0
Feb  4 11:34:44 kooz1 kernel: [100171.512723]  [<ffffffff9990181a>] cpu_startup_entry+0x14a/0x1e0
Feb  4 11:34:44 kooz1 kernel: [100171.512730]  [<ffffffff9985a857>] start_secondary+0x1f7/0x270
Feb  4 11:34:44 kooz1 kernel: [100171.512735]  [<ffffffff998000d5>] start_cpu+0x5/0x14
Feb  4 11:34:44 kooz1 kernel: [100171.512740] ---[ end trace b6616199bcb97c86 ]---
Feb  4 11:34:44 kooz1 kernel: [100171.512846] e1000e 0000:00:19.0 eno1: Reset adapter unexpectedly
Feb  4 11:34:49 kooz1 kernel: [100177.101993] e1000e: eno1 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: Rx/Tx

Et d'après quelques recherches sur Internet, cela peut être sans importance comme cela peut être grave...
Dans quelle direction chercher ?
D'avance merci.

Offline krisden

  • *
  • 43
  • +0/-0
Re: perf: interrupt took too long
« Reply #1 on: February 06, 2022, 12:50:09 PM »
Sur du Proxmox ? D'ailleurs quelles recherches sur Internet ?

Là où tu dois chercher, c'est plutôt :
Code: [Select]
e1000e 0000:00:19.0 enp0s25: Reset adapter unexpectedly
Dans ton autre message, je t'ai répondu que le souci pouvait être résolu à l'aide de "ethtool" (encore faut-il que le problème soit identifié)
Essaye d'apporter un peu d'eau au moulin stp   :-)

Offline gieres

  • *
  • 213
  • +0/-0
Re: perf: interrupt took too long
« Reply #2 on: February 06, 2022, 04:02:03 PM »
Bonjour,
Merci pour les observations.
Quote
Sur du Proxmox ?
Non, pas du tout. Qu'est-ce qui fait penser à cela ?
Quote
D'ailleurs quelles recherches sur Internet ?
Je n'ai cherché que sur la première ligne, pensant que c'était la source du problème :
https://forums.linuxmint.com/viewtopic.php?t=318688
https://forums-linuxmint-com.translate.goog/viewtopic.php?t=318688&_x_tr_sl=de&_x_tr_tl=fr&_x_tr_hl=fr&_x_tr_pto=wapp
https://bugzilla-redhat-com.translate.goog/show_bug.cgi?id=1013708&_x_tr_sl=en&_x_tr_tl=fr&_x_tr_hl=fr&_x_tr_pto=wapp
https://itectec-com.translate.goog/ubuntu/ubuntu-what-does-perf-interupt-mean/?_x_tr_sl=en&_x_tr_tl=fr&_x_tr_hl=fr&_x_tr_pto=wapp

Merci pour l'indication, je vais chercher du côté des réseaux.



Offline gieres

  • *
  • 213
  • +0/-0
[SOLVED] perf: interrupt took too long [SOLVED]
« Reply #3 on: February 13, 2022, 08:20:11 PM »
Donc voir l'autre fil pour les solutions :
https://forums.koozali.org/index.php/topic,54765.0.html
Et on clôt celui-ci.
« Last Edit: February 14, 2022, 06:22:43 PM by gieres »