[squid-users] squid hangs and dies and can not be killed - needs system reboot

Amish anon.amish at gmail.com
Thu Dec 21 12:59:59 UTC 2023


On 21/12/23 17:55, Francesco Chemolli wrote:
> Hi Amish,
>   the message you posted really looks like a kernel bug, possibly due 
> to faulty code, or resulting from a hardware problem.
> Nothing squid can do can cause that kind of stack traces in kernel-space.
>
> A quick search online results in 
> https://lkml.kernel.org/netdev/20230526163458.2880232-1-edumazet@google.com/T/ 
> , which looks very much like your message
>
Yes even I believe so but I wonder, why only after updating to squid 6.6?

Regards

Amish.

> On Thu, Dec 21, 2023 at 1:10 PM Amish <anon.amish at gmail.com> wrote:
>
>     Hi Alex,
>
>     Sorry for late reply. But this is a production system and hence
>     debugging is tough. May take few days.
>
>     However I noticed that everytime squid hangs (goes dead), the
>     dmesg has
>     these errors.
>
>     [63567.802188] divide error: 0000 [#1] PREEMPT SMP PTI
>     [63567.802215] CPU: 3 PID: 617 Comm: squid Not tainted
>     6.6.7-arch1-1 #1
>     4505c4baa0b3d7c4037b0e8f5402626fa360717f
>     [63567.802238] Hardware name: Gigabyte Technology Co., Ltd.
>     H81M-S/H81M-S, BIOS F2 08/19/2015
>     [63567.802255] RIP: 0010:tcp_rcv_space_adjust+0xbe/0x160
>     [63567.802267] Code: f8 41 89 d0 29 d0 31 d2 49 0f af c2 49 f7 f0
>     45 8b
>     81 bc 04 00 00 44 0f b6 8b 10 06 00 00 31 d2 49 8d 04 42 48 98 48
>     c1 e0
>     08 <49> f7 f1 49 63 d0
>     48 98 48 39 d0 48 0f 47 c2 39 83 18 01 00 00 7c
>     [63567.802287] RSP: 0018:ffffba9f00da3c18 EFLAGS: 00010206
>     [63567.802296] RAX: 0000000004fb2800 RBX: ffff9e124acc1c80 RCX:
>     0000000eccd9dace
>     [63567.802304] RDX: 0000000000000000 RSI: 0000000037fa4ae8 RDI:
>     0000000000008349
>     [63567.802314] RBP: ffff9e124acc1c80 R08: 0000000000600000 R09:
>     0000000000000000
>     [63567.802322] R10: 00000000000161d2 R11: ffff9e1346621700 R12:
>     0000000000000000
>     [63567.802331] R13: ffff9e124acc1d58 R14: 0000000000000000 R15:
>     ffff9e124acc2214
>     [63567.802340] FS:  00007fc3d627c840(0000) GS:ffff9e1457380000(0000)
>     knlGS:0000000000000000
>     [63567.802350] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>     [63567.802358] CR2: 0000560b7e86d690 CR3: 0000000102f44001 CR4:
>     00000000001706e0
>     [63567.802367] Call Trace:
>     [63567.802372]  <TASK>
>     [63567.802378]  ? die+0x36/0x90
>     [63567.802386]  ? do_trap+0xda/0x100
>     [63567.802392]  ? tcp_rcv_space_adjust+0xbe/0x160
>     [63567.802401]  ? do_error_trap+0x6a/0x90
>     [63567.802408]  ? tcp_rcv_space_adjust+0xbe/0x160
>     [63567.802415]  ? exc_divide_error+0x38/0x50
>     [63567.802423]  ? tcp_rcv_space_adjust+0xbe/0x160
>     [63567.802431]  ? asm_exc_divide_error+0x1a/0x20
>     [63567.802441]  ? tcp_rcv_space_adjust+0xbe/0x160
>     [63567.802449]  ? tcp_rcv_space_adjust+0x1a/0x160
>     [63567.802456]  tcp_recvmsg_locked+0x2d4/0x960
>     [63567.802464]  tcp_recvmsg+0x87/0x1f0
>     [63567.802471]  inet6_recvmsg+0x56/0x130
>     [63567.802479]  ? __pfx_bpf_lsm_socket_recvmsg+0x10/0x10
>     [63567.802488]  ? security_socket_recvmsg+0x44/0x70
>     [63567.802497]  sock_recvmsg+0x57/0xd0
>     [63567.802505]  sock_read_iter+0x96/0x100
>     [63567.802513]  vfs_read+0x303/0x350
>     [63567.802796]  ksys_read+0xbb/0xf0
>     [63567.803074]  do_syscall_64+0x60/0x90
>     [63567.803346]  ? syscall_exit_to_user_mode+0x2b/0x40
>     [63567.803619]  ? do_syscall_64+0x6c/0x90
>     [63567.803890]  ? syscall_exit_to_user_mode+0x2b/0x40
>     [63567.804163]  ? do_syscall_64+0x6c/0x90
>     [63567.804438]  ? do_syscall_64+0x6c/0x90
>     [63567.804710]  entry_SYSCALL_64_after_hwframe+0x6e/0xd8
>     [63567.804984] RIP: 0033:0x7fc3d5f21531
>     [63567.805271] Code: ff ff eb c3 67 e8 2f c9 01 00 66 2e 0f 1f 84
>     00 00
>     00 00 00 0f 1f 44 00 00 f3 0f 1e fa 80 3d 35 ce 0d 00 00 74 13 31
>     c0 0f
>     05 <48> 3d 00 f0 ff ff
>     77 57 c3 66 0f 1f 44 00 00 48 83 ec 28 48 89 54
>     [63567.805909] RSP: 002b:00007fffa66f6748 EFLAGS: 00000246 ORIG_RAX:
>     0000000000000000
>     [63567.806225] RAX: ffffffffffffffda RBX: 00007fffa66f6830 RCX:
>     00007fc3d5f21531
>     [63567.806542] RDX: 000000000000191a RSI: 000055d106003d80 RDI:
>     00000000000000bd
>     [63567.806855] RBP: 000055d10620bc30 R08: 000055d10620bc30 R09:
>     000055d1036a8e78
>     [63567.807163] R10: 000055d1037ad630 R11: 0000000000000246 R12:
>     000000000000191a
>     [63567.807472] R13: 000055d10664f410 R14: 000055d106003d80 R15:
>     00007fc3d627c6b8
>     [63567.807768]  </TASK>
>     [63567.808037] Modules linked in: nfnetlink_queue
>     nf_conntrack_netlink
>     xt_connmark xt_mark ip_set_hash_net xt_NFQUEUE xt_set ip_set_hash_ip
>     ip_set cls_fw sch_sfq sch_h
>     tb nf_nat_ftp nf_conntrack_ftp tls tun cfg80211 rfkill nft_chain_nat
>     xt_MASQUERADE xt_nat xt_REDIRECT nf_nat nft_limit xt_limit
>     xt_conntrack
>     nf_conntrack nf_defrag_ipv
>     6 nf_defrag_ipv4 xt_multiport xt_tcpudp nft_compat nf_tables
>     libcrc32c
>     intel_rapl_msr intel_rapl_common x86_pkg_temp_thermal
>     intel_powerclamp
>     snd_hda_codec_realtek cor
>     etemp snd_hda_codec_generic ledtrig_audio snd_hda_codec_hdmi
>     kvm_intel
>     snd_hda_intel kvm snd_intel_dspcfg snd_intel_sdw_acpi irqbypass
>     snd_hda_codec snd_hda_core crct1
>     0dif_pclmul r8169 crc32_pclmul snd_hwdep polyval_clmulni snd_pcm
>     polyval_generic gf128mul snd_timer realtek ax88179_178a spi_nor
>     mdio_devres usbnet snd libphy ghash_cl
>     mulni_intel mtd sha512_ssse3 at24 soundcore mii mei_pxp mei_hdcp
>     sha256_ssse3 sha1_ssse3 iTCO_wdt i2c_i801 mei_me spi_intel_platform
>     mousedev aesni_intel intel_pmc_bxt
>       spi_intel iTCO_vendor_support
>     [63567.808076]  i2c_smbus mei crypto_simd lpc_ich cryptd rapl
>     intel_cstate intel_uncore psmouse mac_hid pcspkr pkcs8_key_parser
>     fuse
>     dm_mod loop nfnetlink ip_tables x_
>     tables ext4 crc32c_generic crc16 mbcache jbd2 i915 serio_raw
>     i2c_algo_bit atkbd drm_buddy libps2 ttm vivaldi_fmap intel_gtt
>     xhci_pci
>     crc32c_intel drm_display_helper xh
>     ci_pci_renesas cec i8042 video serio wmi
>     [63567.811333] ---[ end trace 0000000000000000 ]---
>     [63567.811701] RIP: 0010:tcp_rcv_space_adjust+0xbe/0x160
>     [63567.812069] Code: f8 41 89 d0 29 d0 31 d2 49 0f af c2 49 f7 f0
>     45 8b
>     81 bc 04 00 00 44 0f b6 8b 10 06 00 00 31 d2 49 8d 04 42 48 98 48
>     c1 e0
>     08 <49> f7 f1 49 63 d0
>     48 98 48 39 d0 48 0f 47 c2 39 83 18 01 00 00 7c
>     [63567.812848] RSP: 0018:ffffba9f00da3c18 EFLAGS: 00010206
>     [63567.813252] RAX: 0000000004fb2800 RBX: ffff9e124acc1c80 RCX:
>     0000000eccd9dace
>     [63567.813651] RDX: 0000000000000000 RSI: 0000000037fa4ae8 RDI:
>     0000000000008349
>     [63567.814064] RBP: ffff9e124acc1c80 R08: 0000000000600000 R09:
>     0000000000000000
>     [63567.814470] R10: 00000000000161d2 R11: ffff9e1346621700 R12:
>     0000000000000000
>     [63567.814879] R13: ffff9e124acc1d58 R14: 0000000000000000 R15:
>     ffff9e124acc2214
>     [63567.815290] FS:  00007fc3d627c840(0000) GS:ffff9e1457380000(0000)
>     knlGS:0000000000000000
>     [63567.815700] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>     [63567.816121] CR2: 0000560b7e86d690 CR3: 0000000102f44001 CR4:
>     00000000001706e0
>
>     After this the "child" squid process can not be killed, even by
>     systemd,
>     even by kill -9.
>
>     It appears that squid "main" process (owned by root user) does not
>     hang
>     and hence connections on port 8080 are accepted.
>
>     But squid child process (owned by proxy user) hangs and never
>     recovers
>     after above dmesg. And hence nothing happens and connection times out.
>
>     # ps aux |grep squid
>     root         612  0.0  0.2  75500 23680 ?        Ss   Dec20 0:02
>     /usr/bin/squid -f /etc/squid/btnet/squid.btnet.conf --foreground -sYC
>     proxy        617  0.0  0.0      0     0 ?        D    Dec20 1:00
>     [squid]
>     proxy        620  0.0  0.0  12152  7552 ?        S    Dec20 0:00
>     (security_file_certgen) -s /var/cache/squid/ssl_db -M 4MB
>     proxy        621  0.0  0.0  12152  7552 ?        S    Dec20 0:00
>     (security_file_certgen) -s /var/cache/squid/ssl_db -M 4MB
>     proxy        622  0.0  0.0  11888  5504 ?        S    Dec20 0:00
>     (security_file_certgen) -s /var/cache/squid/ssl_db -M 4MB
>     proxy        623  0.0  0.0  11888  5632 ?        S    Dec20 0:00
>     (security_file_certgen) -s /var/cache/squid/ssl_db -M 4MB
>     proxy        624  0.0  0.0  11888  5632 ?        S    Dec20 0:00
>     (security_file_certgen) -s /var/cache/squid/ssl_db -M 4MB
>     proxy        643  0.0  0.0   6116  3200 ?        S    Dec20 0:01
>     (logfile-daemon) /var/log/squid/access.log
>
>     # telnet 127.0.0.1 8080
>     Trying 127.0.0.1...
>     Connected to 127.0.0.1.
>     Escape character is '^]'.
>     ^]
>     telnet> c
>
>     # curl -x 127.0.0.1:8080 <http://127.0.0.1:8080> --max-time 30
>     https://google.com
>     curl: (28) Connection timed out after 30002 milliseconds
>
>     # kill 612 617
>
>     Ran kill 5 times, nothing happened. All squid processes remained.
>
>     # kill -9 612 617
>
>     Ran once, and all squid processes except 617 (child squid process)
>     got
>     killed
>
>     # ps aux |grep squid
>     proxy        617  0.0  0.0      0     0 ?        D    Dec20 1:00
>     [squid]
>     root       90764  0.0  0.0   6552  2560 pts/5    S+   17:36 0:00 grep
>     --color=auto squid
>
>     Can above information be of any help?
>
>     Thanks and regards,
>
>     Amish
>
>     On 19/12/23 20:46, Alex Rousskov wrote:
>     > On 2023-12-18 22:29, Amish wrote:
>     >> On 19/12/23 01:14, Alex Rousskov wrote:
>     >>> On 2023-12-18 09:35, Amish wrote:
>     >>>
>     >>>> I use Arch Linux and today I updated squid from squid 5.7 to
>     squid
>     >>>> 6.6.
>     >>>
>     >>> > Dec 18 13:01:24 mumbai squid[604]: kick abandoning conn199
>     >>>
>     >>> I do not know whether the above problem is the primary problem in
>     >>> your setup, but it is a red flag. Transactions on the same
>     >>> connection may get stuck after that message; it is essentially a
>     >>> Squid bug.
>     >>>
>     >>> I am not sure at all, but this bug might be related to Bug 5187
>     >>> workaround that went into Squid v6.2 (commit c44cfe7):
>     >>> https://bugs.squid-cache.org/show_bug.cgi?id=5187
>     >>>
>     >>> Does Squid accept new TCP connections after it enters what you
>     >>> describe as a dead state? For example, does "telnet 127.0.0.1
>     8080"
>     >>> establishes a connection if executed on the same machine as Squid?
>     >>>
>     >> Yes it establishes connection. But I do not know what to do next.
>     >
>     > This tells us that your Squid is still listening for incoming
>     > connections. Most likely, it is not "dead" but running and just
>     unable
>     > to make progress with those connections (for yet unknown reasons).
>     > That information is helpful but not sufficient (for me) to solve
>     the
>     > problem you are describing.
>     >
>     > The next step that I would recommend is to collect debugging
>     > information from the running process and share a pointer to the
>     > corresponding compressed cache.log file:
>     >
>     > * Ideally, start collection when Squid starts and reproduce the
>     > problem while collecting full debugging information:
>     > http://wiki.squid-cache.org/SquidFaq/BugReporting#full-debug-output
>     >
>     > * If you have to, start collection after Squid is already in bad
>     state
>     > and just before you use telnet or browser to tickle Squid:
>     >
>     http://wiki.squid-cache.org/SquidFaq/BugReporting#debugging-a-single-transaction
>
>     >
>     >
>     > Do not use any secret information (e.g., production certificate
>     keys)
>     > for these tests (unless you are going to share the logs
>     privately with
>     > those you trust).
>     >
>     > Do not downgrade to v5 for these tests.
>     >
>     >
>     > HTH,
>     >
>     > Alex.
>     >
>     >
>     >> Browser showed "Connection timed out" message. But I believe
>     >> browser's also connected but nothing happened afterwards.
>     >>
>     >>>
>     >>> > kill -9 does nothing
>     >>>
>     >>> Is it possible that you are trying to kill the wrong process? You
>     >>> should be killing this process AFAICT:
>     >>>
>     >>> > root         601  0.0  0.2  73816 22528 ?        Ss 12:59 0:02
>     >>> > /usr/bin/squid -f /etc/squid/btnet/squid.btnet.conf
>     --foreground -sYC
>     >>
>     >> I did not clarify but all processes needed SIGKILL and vanished
>     >> except the Dead squid process which still remained.
>     >>
>     >> # systemctl stop squid
>     >>
>     >> Dec 19 08:46:38 mumbai systemd[1]: squid.service: State
>     >> 'stop-sigterm' timed out. Killing.
>     >> Dec 19 08:46:38 mumbai systemd[1]: squid.service: Killing
>     process 601
>     >> (squid) with signal SIGKILL.
>     >> Dec 19 08:46:38 mumbai systemd[1]: squid.service: Killing
>     process 604
>     >> (squid) with signal SIGKILL.
>     >> Dec 19 08:46:38 mumbai systemd[1]: squid.service: Killing
>     process 607
>     >> (security_file_c) with signal SIGKILL.
>     >> Dec 19 08:46:38 mumbai systemd[1]: squid.service: Killing
>     process 608
>     >> (security_file_c) with signal SIGKILL.
>     >> Dec 19 08:46:38 mumbai systemd[1]: squid.service: Killing
>     process 609
>     >> (security_file_c) with signal SIGKILL.
>     >> Dec 19 08:46:38 mumbai systemd[1]: squid.service: Killing
>     process 610
>     >> (security_file_c) with signal SIGKILL.
>     >> Dec 19 08:46:38 mumbai systemd[1]: squid.service: Killing
>     process 611
>     >> (security_file_c) with signal SIGKILL.
>     >> Dec 19 08:46:38 mumbai systemd[1]: squid.service: Killing
>     process 622
>     >> (log_file_daemon) with signal SIGKILL.
>     >> Dec 19 08:46:38 mumbai systemd[1]: squid.service: Main process
>     >> exited, code=killed, status=9/KILL
>     >> Dec 19 08:46:38 mumbai systemd[1]: squid.service: Killing
>     process 604
>     >> (squid) with signal SIGKILL.
>     >>
>     >> Waited for 2 minutes for squid to stop then pressed ctrl-c to
>     >> systemctl stop squid command.
>     >>
>     >> As you can see in last line shows that attempt was made to kill
>     DEAD
>     >> process with PID 604.
>     >>
>     >> # ps aux |grep squid
>     >> proxy        604  0.0  0.0      0     0 ?        D Dec18 0:03
>     [squid]
>     >>
>     >> Now only DEAD squid process remains.
>     >>
>     >> What next? Should I downgrade to 5.9 and check?
>     >>
>     >> Regards
>     >>
>     >> Amish
>     >>
>     >>> Alex.
>     >>>
>     >>>
>     >>>> After the update from 5.7 to 6.6, squid starts but then reaches
>     >>>> Dead state in a minute or two.
>     >>>>
>     >>>> # ps aux | grep squid
>     >>>> root         601  0.0  0.2  73816 22528 ?        Ss   12:59 0:02
>     >>>> /usr/bin/squid -f /etc/squid/btnet/squid.btnet.conf
>     --foreground -sYC
>     >>>> proxy        604  0.0  0.0      0     0 ?        D    12:59 0:03
>     >>>> [squid]
>     >>>> proxy        607  0.0  0.0  11976  7424 ?        S    12:59 0:00
>     >>>> (security_file_certgen) -s /var/cache/squid/ssl_db -M 4MB
>     >>>> proxy        608  0.0  0.0  11976  7168 ?        S    12:59 0:00
>     >>>> (security_file_certgen) -s /var/cache/squid/ssl_db -M 4MB
>     >>>> proxy        609  0.0  0.0  11712  5632 ?        S    12:59 0:00
>     >>>> (security_file_certgen) -s /var/cache/squid/ssl_db -M 4MB
>     >>>> proxy        610  0.0  0.0  11712  5376 ?        S    12:59 0:00
>     >>>> (security_file_certgen) -s /var/cache/squid/ssl_db -M 4MB
>     >>>> proxy        611  0.0  0.0  11712  5504 ?        S    12:59 0:00
>     >>>> (security_file_certgen) -s /var/cache/squid/ssl_db -M 4MB
>     >>>> proxy        622  0.0  0.0   6116  3200 ?        S    12:59 0:00
>     >>>> (logfile-daemon) /var/log/squid/access.log
>     >>>>
>     >>>> And then all requests get stuck. Notice the D (dead) state of
>     squid.
>     >>>>
>     >>>> I use multiple ports for multiple purposes. (It all worked
>     fine in
>     >>>> squid 5.7)
>     >>>>
>     >>>> Dec 18 12:59:10 mumbai squid[601]: Starting Authentication on
>     port
>     >>>> [::]:3128
>     >>>> Dec 18 12:59:10 mumbai squid[601]: Disabling Authentication
>     on port
>     >>>> [::]:3128 (interception enabled)
>     >>>> Dec 18 12:59:10 mumbai squid[601]: Starting Authentication on
>     port
>     >>>> [::]:8081
>     >>>> Dec 18 12:59:10 mumbai squid[601]: Disabling Authentication
>     on port
>     >>>> [::]:8081 (interception enabled)
>     >>>> Dec 18 12:59:12 mumbai squid[601]: Starting Authentication on
>     port
>     >>>> [::]:8082
>     >>>> Dec 18 12:59:12 mumbai squid[601]: Disabling Authentication
>     on port
>     >>>> [::]:8082 (interception enabled)
>     >>>> Dec 18 12:59:12 mumbai squid[601]: Starting Authentication on
>     port
>     >>>> [::]:8083
>     >>>> Dec 18 12:59:12 mumbai squid[601]: Disabling Authentication
>     on port
>     >>>> [::]:8083 (interception enabled)
>     >>>> Dec 18 12:59:13 mumbai squid[601]: Starting Authentication on
>     port
>     >>>> [::]:8084
>     >>>> Dec 18 12:59:13 mumbai squid[601]: Disabling Authentication
>     on port
>     >>>> [::]:8084 (interception enabled)
>     >>>> Dec 18 12:59:13 mumbai squid[601]: Starting Authentication on
>     port
>     >>>> [::]:3136
>     >>>> Dec 18 12:59:13 mumbai squid[601]: Disabling Authentication
>     on port
>     >>>> [::]:3136 (interception enabled)
>     >>>> Dec 18 12:59:13 mumbai squid[601]: Starting Authentication on
>     port
>     >>>> [::]:3137
>     >>>> Dec 18 12:59:13 mumbai squid[601]: Disabling Authentication
>     on port
>     >>>> [::]:3137 (interception enabled)
>     >>>> ...
>     >>>> Dec 18 12:59:29 mumbai squid[604]: Adaptation support is on
>     >>>> Dec 18 12:59:29 mumbai squid[604]: Accepting NAT intercepted
>     HTTP
>     >>>> Socket connections at conn19 local=[::]:3128 remote=[::] FD 27
>     >>>> flags=41
>     >>>> listening port: 3128
>     >>>> Dec 18 12:59:29 mumbai squid[604]: Accepting SSL bumped HTTP
>     Socket
>     >>>> connections at conn21 local=[::]:8080 remote=[::] FD 28 flags=9
>     >>>> listening port: 8080
>     >>>> Dec 18 12:59:29 mumbai squid[604]: Accepting NAT intercepted SSL
>     >>>> bumped HTTPS Socket connections at conn23 local=[::]:8081
>     >>>> remote=[::] FD 29 flags=41
>     >>>> listening port: 8081
>     >>>> Dec 18 12:59:29 mumbai squid[604]: Accepting SSL bumped HTTP
>     Socket
>     >>>> connections at conn25 local=[::]:8092 remote=[::] FD 30 flags=9
>     >>>> listening port: 8092
>     >>>> Dec 18 12:59:29 mumbai systemd[1]: Started Squid Web Proxy
>     Server.
>     >>>> Dec 18 12:59:29 mumbai squid[604]: Accepting SSL bumped HTTP
>     Socket
>     >>>> connections at conn27 local=[::]:8093 remote=[::] FD 31 flags=9
>     >>>> listening port: 8093
>     >>>> Dec 18 12:59:29 mumbai squid[604]: Accepting SSL bumped HTTP
>     Socket
>     >>>> connections at conn29 local=[::]:8094 remote=[::] FD 32 flags=9
>     >>>> listening port: 8094
>     >>>> Dec 18 12:59:29 mumbai squid[604]: Accepting NAT intercepted SSL
>     >>>> bumped HTTPS Socket connections at conn31 local=[::]:8082
>     >>>> remote=[::] FD 33 flags=41
>     >>>> listening port: 8082
>     >>>> Dec 18 12:59:29 mumbai squid[604]: Accepting NAT intercepted SSL
>     >>>> bumped HTTPS Socket connections at conn33 local=[::]:8083
>     >>>> remote=[::] FD 34 flags=41
>     >>>> listening port: 8083
>     >>>> Dec 18 12:59:29 mumbai squid[604]: Accepting NAT intercepted SSL
>     >>>> bumped HTTPS Socket connections at conn35 local=[::]:8084
>     >>>> remote=[::] FD 35 flags=41
>     >>>> listening port: 8084
>     >>>> Dec 18 12:59:29 mumbai squid[604]: Accepting NAT intercepted
>     HTTP
>     >>>> Socket connections at conn37 local=[::]:3136 remote=[::] FD 36
>     >>>> flags=41
>     >>>> listening port: 3136
>     >>>> Dec 18 12:59:29 mumbai squid[604]: Accepting NAT intercepted
>     HTTP
>     >>>> Socket connections at conn39 local=[::]:3137 remote=[::] FD 37
>     >>>> flags=41
>     >>>> listening port: 3137
>     >>>>
>     >>>> And then following errors came:
>     >>>>
>     >>>>
>     >>>> Dec 18 12:59:45 mumbai squid[604]: ERROR: failure while
>     accepting a
>     >>>> TLS connection on conn41 local=192.168.0.1:8080
>     <http://192.168.0.1:8080>
>     >>>> remote=192.168.0.111:53867 <http://192.168.0.111:53867> FD 12
>     flags=1: SQUID_TLS
>     >>>> _ERR_ACCEPT+TLS_LIB_ERR=A000416+TLS_IO_ERR=1
>     >>>> current master transaction:
>     >>>> master53
>     >>>> Dec 18 12:59:45 mumbai squid[604]: ERROR: failure while
>     accepting a
>     >>>> TLS connection on conn42 local=192.168.0.1:8080
>     <http://192.168.0.1:8080>
>     >>>> remote=192.168.0.111:53868 <http://192.168.0.111:53868> FD 14
>     flags=1: SQUID_TLS
>     >>>> _ERR_ACCEPT+TLS_LIB_ERR=A000416+TLS_IO_ERR=1
>     >>>> current master transaction:
>     >>>> master53
>     >>>> Dec 18 12:59:45 mumbai squid[604]: ERROR: failure while
>     accepting a
>     >>>> TLS connection on conn43 local=192.168.0.1:8080
>     <http://192.168.0.1:8080>
>     >>>> remote=192.168.0.111:53869 <http://192.168.0.111:53869> FD 16
>     flags=1: SQUID_TLS
>     >>>> _ERR_ACCEPT+TLS_LIB_ERR=A000416+TLS_IO_ERR=1
>     >>>> current master transaction:
>     >>>> master57
>     >>>> Dec 18 12:59:45 mumbai squid[604]: ERROR: failure while
>     accepting a
>     >>>> TLS connection on conn44 local=192.168.0.1:8080
>     <http://192.168.0.1:8080>
>     >>>> remote=192.168.0.111:53870 <http://192.168.0.111:53870> FD 12
>     flags=1: SQUID_TLS
>     >>>> _ERR_ACCEPT+TLS_LIB_ERR=A000416+TLS_IO_ERR=1
>     >>>> current master transaction:
>     >>>> master57
>     >>>> Dec 18 12:59:56 mumbai squid[604]: ERROR: failure while
>     accepting a
>     >>>> TLS connection on conn62 local=192.168.0.1:8080
>     <http://192.168.0.1:8080>
>     >>>> remote=192.168.0.111:53887 <http://192.168.0.111:53887> FD 12
>     flags=1: SQUID_TLS
>     >>>> _ERR_ACCEPT+TLS_LIB_ERR=A000416+TLS_IO_ERR=1
>     >>>> current master transaction:
>     >>>> master95
>     >>>> Dec 18 12:59:59 mumbai squid[604]: ERROR: failure while
>     accepting a
>     >>>> TLS connection on conn64 local=192.168.0.1:8080
>     <http://192.168.0.1:8080>
>     >>>> remote=192.168.0.111:53888 <http://192.168.0.111:53888> FD 12
>     flags=1: SQUID_TLS
>     >>>> _ERR_ACCEPT+TLS_LIB_ERR=A000416+TLS_IO_ERR=1
>     >>>> current master transaction:
>     >>>> master99
>     >>>> Dec 18 13:00:02 mumbai squid[604]: ERROR: failure while
>     accepting a
>     >>>> TLS connection on conn65 local=192.168.0.1:8080
>     <http://192.168.0.1:8080>
>     >>>> remote=192.168.0.178:56115 <http://192.168.0.178:56115> FD 12
>     flags=1: SQUID_TLS
>     >>>> _ERR_ACCEPT+TLS_LIB_ERR=A000418+TLS_IO_ERR=1
>     >>>> current master transaction:
>     >>>> master53
>     >>>> Dec 18 13:01:24 mumbai squid[604]: kick abandoning conn199
>     >>>> local=192.168.0.1:8093 <http://192.168.0.1:8093>
>     remote=192.168.0.101:52211 <http://192.168.0.101:52211> FD 52 flags=1
>     >>>> connection: conn199
>     >>>> local=192.168.0.1:8093 <http://192.168.0.1:8093>
>     remote=192.168.0.101:52211 <http://192.168.0.101:52211> FD 52 flags=1
>     >>>> Dec 18 13:01:45 mumbai squid[604]: ERROR: failure while
>     accepting a
>     >>>> TLS connection on conn240 local=192.168.0.1:8093
>     <http://192.168.0.1:8093>
>     >>>> remote=192.168.0.111:53931 <http://192.168.0.111:53931> FD 48
>     flags=1: SQUID_TL
>     >>>> S_ERR_ACCEPT+TLS_LIB_ERR=A000416+TLS_IO_ERR=1
>     >>>> current master transaction:
>     >>>> master314
>     >>>>
>     >>>>
>     >>>> After this point there is nothing in systemd journal (via:
>     >>>> journalctl -f -u squid) and same lines are in cache.log.
>     >>>>
>     >>>> Squid got stuck (DEAD state) at 13:01 and right now it 19:26 (6
>     >>>> hours passed) and squid is still in dead state.
>     >>>>
>     >>>> kill -9 or kill -ALRM or -HUP also does nothing.
>     >>>>
>     >>>> So to restart squid - I will need to restart whole system.
>     >>>>
>     >>>> I have sslbump directives but it is not really applied.
>     >>>>
>     >>>> #NOTE: nosslbump_ips below contains 192.168.0.0/24
>     <http://192.168.0.0/24> (whole LAN) so
>     >>>> effectively there is no SSL bump after step1.
>     >>>>
>     >>>> acl nosslbump_ips src 192.168.0.0/24 <http://192.168.0.0/24>
>     >>>> ssl_bump splice ssl_step1 nosslbump_ips
>     >>>> ssl_bump peek ssl_step1
>     >>>> ssl_bump splice nosslbump_domains
>     >>>> ssl_bump stare sslbump_domains
>     >>>> ssl_bump splice ssl_step2
>     >>>> ssl_bump bump all
>     >>>>
>     >>>>
>     >>>> Any idea? If anything changed from 5.7 to 6.6 that may cause
>     this
>     >>>> behaviour?
>     >>>>
>     >>>> Looking at changelog:
>     >>>>
>     >>>> Bug 5256: Intercepting port fails to accept
>     >>>> https://bugs.squid-cache.org/show_bug.cgi?id=5256
>     >>>>
>     >>>> Bug 5154: Do not open IPv6 sockets when IPv6 is disabled
>     >>>> https://bugs.squid-cache.org/show_bug.cgi?id=5154
>     >>>>
>     >>>> Not sure if above two bug FIXES (in between v5.7 to v6.6) are
>     >>>> related to my issue.
>     >>>>
>     >>>> I ran netstat:
>     >>>>
>     >>>> # netstat -ntlp
>     >>>> Active Internet connections (only servers)
>     >>>> Proto Recv-Q Send-Q Local Address Foreign Address
>     >>>> State       PID/Program name
>     >>>> ...
>     >>>> tcp6      33      0 :::3137 :::*                    LISTEN -
>     >>>> tcp6       0      0 :::3136 :::*                    LISTEN -
>     >>>> tcp6       4      0 :::3128 :::*                    LISTEN -
>     >>>> tcp6       0      0 :::8081 :::*                    LISTEN -
>     >>>> tcp6       0      0 :::8080 :::*                    LISTEN -
>     >>>> tcp6       0      0 :::8083 :::*                    LISTEN -
>     >>>> tcp6       0      0 :::8082 :::*                    LISTEN -
>     >>>> tcp6       0      0 :::8084 :::*                    LISTEN -
>     >>>> tcp6    4097      0 :::8093 :::*                    LISTEN -
>     >>>> tcp6       0      0 :::8092 :::*                    LISTEN -
>     >>>> tcp6       0      0 :::8094 :::*                    LISTEN -
>     >>>> ...
>     >>>>
>     >>>> I do not have IPv6 enabled, yet there are 33 and 4097 numbers in
>     >>>> Recv-Q and also no process/PID owns these ports.
>     >>>>
>     >>>> Same IPv4 ports are not shown in use by netstat, so only IPv6
>     ports
>     >>>> remain open, that too orphaned!
>     >>>>
>     >>>> So what is happening?
>     >>>>
>     >>>> Any idea to solve or any workaround?
>     >>>>
>     >>>> Thank you,
>     >>>>
>     >>>> Amish.
>     >> _______________________________________________
>     >> squid-users mailing list
>     >> squid-users at lists.squid-cache.org
>     >> https://lists.squid-cache.org/listinfo/squid-users
>     >
>     _______________________________________________
>     squid-users mailing list
>     squid-users at lists.squid-cache.org
>     https://lists.squid-cache.org/listinfo/squid-users
>
>
>
> -- 
>     Francesco
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.squid-cache.org/pipermail/squid-users/attachments/20231221/a1b1a966/attachment-0001.htm>


More information about the squid-users mailing list