eve-comm on GCP: kernel crash from uksm

Before posting something, READ the changelog, WATCH the videos, howto and provide following:
Your install is: Bare metal, ESXi, what CPU model, RAM, HD, what EVE version you have, output of the uname -a and any other info that might help us faster.

Moderator: mike

Post Reply
ymeillier
Posts: 10
Joined: Mon Feb 03, 2020 1:30 am

eve-comm on GCP: kernel crash from uksm

Post by ymeillier » Tue Dec 06, 2022 3:36 pm

Hi.

I'm running eve-ng community on GCP.

eve-ng breaks consistently once i start powering up a few nodes in my topology (vyos nodes with 512 MB ram. when it crashes is random as far as how mayn nodes have been powered up).

The logs seem to point to issues with memory sharing issues with uksm. See the following kernel trace:

-- Trace ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Dec 6 07:04:13 eve-ng kernel: [ 245.991835] WARNING: CPU: 7 PID: 114 at arch/x86/kvm/../../../virt/kvm/kvm_main.c:649 kvm_mmu_notifier_change_pte+0x28a/0x2b0 [kvm]
Dec 6 07:04:13 eve-ng kernel: [ 245.992012] Modules linked in: bridge stp llc nls_iso8859_1 dm_multipath scsi_dh_rdac scsi_dh_emc scsi_dh_alua intel_rapl_msr intel_rapl_common isst_if_common nfit kvm_intel binfmt_misc kvm rapl input_leds serio_raw mac_hid sch_fq_codel ramoops reed_solomon efi_pstore drm virtio_rng ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear crct10dif_pclmul crc32_pclmul ghash_clmulni_intel aesni_intel virtio_net net_failover crypto_simd cryptd psmouse virtio_scsi failover i2c_piix4
Dec 6 07:04:13 eve-ng kernel: [ 245.992092] CPU: 7 PID: 114 Comm: uksmd Not tainted 5.17.8-eve-ng-uksm-wg+ #1
Dec 6 07:04:13 eve-ng kernel: [ 245.992095] Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 10/11/2022
Dec 6 07:04:13 eve-ng kernel: [ 245.992097] RIP: 0010:kvm_mmu_notifier_change_pte+0x28a/0x2b0 [kvm]
Dec 6 07:04:13 eve-ng kernel: [ 245.992140] Code: ff 4c 89 ef e8 67 25 20 cc e9 5b ff ff ff 4c 89 ef 44 88 4d a0 e8 36 fd ff ff 44 0f b6 4d a0 45 84 c9 0f 84 41 ff ff ff eb d7 <0f> 0b e9 b8 fd ff ff 0f 0b e9 4d ff ff ff 0f 0b e9 37 ff ff ff e8
Dec 6 07:04:13 eve-ng kernel: [ 245.992143] RSP: 0018:ffff9d1dc0653bb0 EFLAGS: 00010246
Dec 6 07:04:13 eve-ng kernel: [ 245.992145] RAX: 0000000000000000 RBX: ffff9d1dc1e62c18 RCX: 800000017a157807
Dec 6 07:04:13 eve-ng kernel: [ 245.992146] RDX: 00007f3da1f57000 RSI: ffff89208fb2f2c0 RDI: ffff9d1dc1e62c18
Dec 6 07:04:13 eve-ng kernel: [ 245.992148] RBP: ffff9d1dc0653c38 R08: 0000000000000000 R09: 0000000000000000
Dec 6 07:04:13 eve-ng kernel: [ 245.992149] R10: 0000000000000000 R11: 0000000000000000 R12: ffff89208fb2f2c0
Dec 6 07:04:13 eve-ng kernel: [ 245.992150] R13: 800000017a157807 R14: 00007f3da1f57000 R15: 00007f3da1f57000
Dec 6 07:04:13 eve-ng kernel: [ 245.992152] FS: 0000000000000000(0000) GS:ffff89279fdc0000(0000) knlGS:0000000000000000
Dec 6 07:04:13 eve-ng kernel: [ 245.992154] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Dec 6 07:04:13 eve-ng kernel: [ 245.992155] CR2: 000055fe42609a08 CR3: 000000010f238006 CR4: 00000000003726e0
Dec 6 07:04:13 eve-ng kernel: [ 245.992159] Call Trace:
Dec 6 07:04:13 eve-ng kernel: [ 245.992161] <TASK>
Dec 6 07:04:13 eve-ng kernel: [ 245.992166] ? kvm_arch_mmu_notifier_invalidate_range+0x21/0x50 [kvm]
Dec 6 07:04:13 eve-ng kernel: [ 245.992215] __mmu_notifier_change_pte+0x55/0x90
Dec 6 07:04:13 eve-ng kernel: [ 245.992221] restore_uksm_page_pte+0x23e/0x250
Dec 6 07:04:13 eve-ng kernel: [ 245.992225] cmp_and_merge_page+0x89c/0x2580
Dec 6 07:04:13 eve-ng kernel: [ 245.992229] scan_vma_one_page+0x500/0x1720
Dec 6 07:04:13 eve-ng kernel: [ 245.992233] uksm_do_scan+0x15a/0x3130
Dec 6 07:04:13 eve-ng kernel: [ 245.992236] ? del_timer_sync+0x29/0x40
Dec 6 07:04:13 eve-ng kernel: [ 245.992240] ? schedule_timeout+0x19c/0x290
Dec 6 07:04:13 eve-ng kernel: [ 245.992244] ? uksm_do_scan+0x3130/0x3130
Dec 6 07:04:13 eve-ng kernel: [ 245.992246] uksm_scan_thread+0x164/0x1a0
Dec 6 07:04:13 eve-ng kernel: [ 245.992249] ? _raw_spin_unlock_irqrestore+0x29/0x3d
Dec 6 07:04:13 eve-ng kernel: [ 245.992252] kthread+0xfd/0x130
Dec 6 07:04:13 eve-ng kernel: [ 245.992255] ? kthread_complete_and_exit+0x20/0x20
Dec 6 07:04:13 eve-ng kernel: [ 245.992259] ret_from_fork+0x1f/0x30
Dec 6 07:04:13 eve-ng kernel: [ 245.992264] </TASK>
Dec 6 07:04:13 eve-ng kernel: [ 245.992265] ---[ end trace 0000000000000000 ]---


-- Eve-ng system details: ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

$ eve-info
Tue Dec 6 07:01:49 MST 2022
---------------Packages Installed----------------
ii eve-ng 5.0.1-13
ii eve-ng-dynamips 5.0.1-1
ii eve-ng-pro-guacamole 5.0.1-1
ii eve-ng-qemu 5.0.1-1
ii eve-ng-schema 5.0.1-1
ii eve-ng-vpcs 5.1-eve-ng
ii linux-image-5.17.8-eve-ng-uksm-wg+ 1

---------------Hostname--------------------------
Static hostname: eve-ng
Virtualization: kvm
Operating System: Ubuntu 20.04.5 LTS
Kernel: Linux 5.17.8-eve-ng-uksm-wg+
Architecture: x86-64
---------------Disk Usage------------------------
Filesystem Size Used Avail Use% Mounted on
udev 16G 0 16G 0% /dev
tmpfs 3.2G 1.2M 3.2G 1% /run
/dev/sda1 49G 13G 36G 26% /
tmpfs 16G 0 16G 0% /dev/shm
tmpfs 5.0M 0 5.0M 0% /run/lock
tmpfs 16G 0 16G 0% /sys/fs/cgroup
/dev/loop0 56M 56M 0 100% /snap/core18/2566
/dev/loop1 56M 56M 0 100% /snap/core18/2632
/dev/loop2 64M 64M 0 100% /snap/core20/1695
/dev/loop3 64M 64M 0 100% /snap/core20/1623
/dev/loop4 303M 303M 0 100% /snap/google-cloud-cli/89
/dev/loop5 92M 92M 0 100% /snap/lxd/23991
/dev/loop6 302M 302M 0 100% /snap/google-cloud-cli/77
/dev/sda15 105M 5.2M 100M 5% /boot/efi
/dev/loop7 92M 92M 0 100% /snap/lxd/24061
/dev/loop8 50M 50M 0 100% /snap/snapd/17883
tmpfs 3.2G 0 3.2G 0% /run/user/640247063

---------------CPU Info--------------------------
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Byte Order: Little Endian
Address sizes: 46 bits physical, 48 bits virtual
CPU(s): 8
On-line CPU(s) list: 0-7
Thread(s) per core: 2
Core(s) per socket: 4
Socket(s): 1
NUMA node(s): 1
Vendor ID: GenuineIntel
CPU family: 6
Model: 85
Model name: Intel(R) Xeon(R) CPU @ 2.80GHz
Stepping: 7
CPU MHz: 2800.210
BogoMIPS: 5600.42
Virtualization: VT-x
Hypervisor vendor: KVM
Virtualization type: full
L1d cache: 128 KiB
L1i cache: 128 KiB
L2 cache: 4 MiB
L3 cache: 33 MiB
NUMA node0 CPU(s): 0-7
Vulnerability Itlb multihit: Not affected
Vulnerability L1tf: Not affected
Vulnerability Mds: Mitigation; Clear CPU buffers; SMT Host state unknown
Vulnerability Meltdown: Not affected
Vulnerability Spec store bypass: Vulnerable
Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2: Vulnerable, IBPB: disabled, STIBP: disabled
Vulnerability Srbds: Not affected
Vulnerability Tsx async abort: Mitigation; Clear CPU buffers; SMT Host state unknown
Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss ht syscall nx pdpe1gb rdtscp lm constant_tsc rep_good nopl xtopology nonstop_tsc cpuid tsc_known_freq pni pclmulqdq vmx ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch invpcid_single ssbd ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi
flexpriority ept vpid fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm avx512f avx512dq rdseed adx smap clflushopt clwb avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves arat avx512_vnni md_clear arch_capabilities

---------------Memory Info-----------------------
total used free shared buff/cache available
Mem: 31Gi 968Mi 29Gi 3.0Mi 563Mi 30Gi
Swap: 0B 0B 0B

---------------Nic Info--------------------------
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 8896 qdisc fq_codel master pnet0 state UP mode DEFAULT group default qlen 1000
3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 8896 qdisc fq_codel master pnet1 state UP mode DEFAULT group default qlen 1000
4: eth2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 8896 qdisc fq_codel master pnet2 state UP mode DEFAULT group default qlen 1000
5: eth3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 8896 qdisc fq_codel master pnet3 state UP mode DEFAULT group default qlen 1000
6: eth4: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 8896 qdisc fq_codel master pnet4 state UP mode DEFAULT group default qlen 1000
7: eth5: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 8896 qdisc fq_codel master pnet5 state UP mode DEFAULT group default qlen 1000
8: eth6: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 8896 qdisc fq_codel master pnet6 state UP mode DEFAULT group default qlen 1000
9: eth7: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 8896 qdisc fq_codel master pnet7 state UP mode DEFAULT group default qlen 1000

---------------IP Info---------------------------
● State: routable

---------------Bridge Info-----------------------
pnet0 8000.42010a0a0a0a no eth0
pnet1 8000.42010a0a1414 no eth1
pnet2 8000.42010a0a1e1e no eth2
pnet3 8000.42010a0a2828 no eth3
pnet4 8000.42010a0a3232 no eth4
pnet5 8000.42010a0a3c3c no eth5
pnet6 8000.42010a0a4646 no eth6
pnet7 8000.42010a0a5050 no eth7

---------------H/W Accel-------------------------
INFO: /dev/kvm exists
KVM acceleration can be used

---------------Service Info----------------------
-------------------------------------------------
--------------Guacamole--------------------------
● guacd.service - LSB: Guacamole proxy daemon
Loaded: loaded (/etc/init.d/guacd; generated)
Active: active (running) since Tue 2022-12-06 07:00:20 MST; 1min 29s ago

--------------Tomcat-----------------------------
Unit tomcat8.service could not be found.

--------------Mysql------------------------------
● mysql.service - MySQL Community Server
Loaded: loaded (/lib/systemd/system/mysql.service; enabled; vendor preset: enabled)
Active: active (running) since Tue 2022-12-06 07:00:23 MST; 1min 26s ago

--------------Apache-----------------------------
● apache2.service - The Apache HTTP Server
Loaded: loaded (/lib/systemd/system/apache2.service; enabled; vendor preset: enabled)
Active: active (running) since Tue 2022-12-06 07:00:22 MST; 1min 27s ago



-- GCP GCE Instance details -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
admin_@cloudshell:~ (eve-ng-368801)$ gcloud compute instances describe eve-ng --zone=us-central1-a
canIpForward: true
confidentialInstanceConfig:
enableConfidentialCompute: false
cpuPlatform: Intel Cascade Lake
creationTimestamp: '2022-12-02T10:37:47.937-08:00'
deletionProtection: false
description: ''
disks:
- architecture: X86_64
autoDelete: true
boot: true
deviceName: eve-ng
diskSizeGb: '50'
guestOsFeatures:
- type: VIRTIO_SCSI_MULTIQUEUE
- type: SEV_CAPABLE
- type: UEFI_COMPATIBLE
- type: GVNIC
index: 0
interface: SCSI
kind: compute#attachedDisk
licenses:
- https://www.googleapis.com/compute/v1/p ... u-2004-lts
- https://www.googleapis.com/compute/v1/p ... enable-vmx
mode: READ_WRITE


Not sure if it is an issue with ukms or with GCP nested virtualization implementation/support

Please advise.

Thank you.

ymeillier
Posts: 10
Joined: Mon Feb 03, 2020 1:30 am

Re: eve-comm on GCP: kernel crash from uksm

Post by ymeillier » Thu Dec 08, 2022 12:49 am

Google Engg reviewed the offending trace and commented the following.

"This is a flaw in the L1 kernel, which appears to be running out-of-tree "Ultra KSM". The offending code is from https://github.com/dolohow/uksm/blob/ma ... 5.17.patch. As noted by the commit that added the warn, c13fda237f08 ("KVM: Assert that notifier count is elevated in .change_pte()") the in-tree ->change_pte() hook got unintentionally neutered a while back. The UKSM code is running afoul of KVM's paranoia about how the MMU notifiers are used. UKSM might be able to simply delete the line in KVM, but I would not be remotely confident that UKSM is bug free on recent kernels.
Given that the commit in question is a patch, I think it would be reasonable to treat this as a bug in uksm patch code.
In particular, to fix the problem, they should update line 3131 with mmu_notifier_invalidate_range_{start,end} before and after this line. If uksm wants to improve the mmu_notifier interaction with KVM by avoiding the usage of this wrapper, then sending out an upstream patch to KVM community would be essentially needed."

eve runs fine until multiple nodes are started (vyos in my case).
I tried running eve with ukms disabled but when doing so it 'crashes' (linux network stack is borked) upon starting my first node.

Has eve-comm been tested on GCP recently?

comments?

Uldis (UD)
Posts: 5067
Joined: Wed Mar 15, 2017 4:44 pm
Location: London
Contact:

Re: eve-comm on GCP: kernel crash from uksm

Post by Uldis (UD) » Thu Dec 08, 2022 9:20 am

No idea how you created EVE community on GCP but with vyatta vyos, and other qemu nodes like vIOS no issues
Used EVE-Community cookbook to deploy it.

Screen below from GCP EVE Community lab running
EVE Comm version 5.0.1-13

Code: Select all

root@eve-conn-v5:~# eve-info
Thu Dec  8 09:24:07 UTC 2022
---------------Packages Installed----------------
ii eve-ng 5.0.1-13
ii eve-ng-dynamips 5.0.1-1
ii eve-ng-pro-guacamole 5.0.1-1
ii eve-ng-qemu 5.0.1-1
ii eve-ng-schema 5.0.1-1
ii eve-ng-vpcs 5.1-eve-ng
ii linux-image-5.17.8-eve-ng-uksm-wg+ 1

---------------Hostname--------------------------
   Static hostname: eve-ng
Transient hostname: eve-conn-v5
    Virtualization: kvm
  Operating System: Ubuntu 20.04.5 LTS
            Kernel: Linux 5.17.8-eve-ng-uksm-wg+
      Architecture: x86-64
---------------Disk Usage------------------------
Filesystem      Size  Used Avail Use% Mounted on
udev            7.8G     0  7.8G   0% /dev
tmpfs           1.6G  1.4M  1.6G   1% /run
/dev/sda1        49G  7.0G   42G  15% /
tmpfs           7.9G     0  7.9G   0% /dev/shm
tmpfs           5.0M     0  5.0M   0% /run/lock
tmpfs           7.9G     0  7.9G   0% /sys/fs/cgroup
/dev/loop0       56M   56M     0 100% /snap/core18/2344
/dev/loop1       62M   62M     0 100% /snap/core20/1405
/dev/loop2      295M  295M     0 100% /snap/google-cloud-sdk/235
/dev/loop3       68M   68M     0 100% /snap/lxd/22753
/dev/loop4       44M   44M     0 100% /snap/snapd/15177
/dev/sda15      105M  5.2M  100M   5% /boot/efi
tmpfs           1.6G     0  1.6G   0% /run/user/1001
tmpfs           1.6G     0  1.6G   0% /run/user/0

---------------CPU Info--------------------------
Architecture:                    x86_64
CPU op-mode(s):                  32-bit, 64-bit
Byte Order:                      Little Endian
Address sizes:                   46 bits physical, 48 bits virtual
CPU(s):                          4
On-line CPU(s) list:             0-3
Thread(s) per core:              2
Core(s) per socket:              2
Socket(s):                       1
NUMA node(s):                    1
Vendor ID:                       GenuineIntel
CPU family:                      6
Model:                           85
Model name:                      Intel(R) Xeon(R) CPU @ 2.80GHz
Stepping:                        7
CPU MHz:                         2800.216
BogoMIPS:                        5600.43
Virtualization:                  VT-x
Hypervisor vendor:               KVM
Virtualization type:             full
L1d cache:                       64 KiB
L1i cache:                       64 KiB
L2 cache:                        2 MiB
L3 cache:                        33 MiB
NUMA node0 CPU(s):               0-3
Vulnerability Itlb multihit:     Not affected
Vulnerability L1tf:              Not affected
Vulnerability Mds:               Mitigation; Clear CPU buffers; SMT Host st
                                 ate unknown
Vulnerability Meltdown:          Not affected
Vulnerability Spec store bypass: Vulnerable
Vulnerability Spectre v1:        Mitigation; usercopy/swapgs barriers and _
                                 _user pointer sanitization
Vulnerability Spectre v2:        Vulnerable, IBPB: disabled, STIBP: disable
                                 d
Vulnerability Srbds:             Not affected
Vulnerability Tsx async abort:   Mitigation; Clear CPU buffers; SMT Host st
                                 ate unknown
Flags:                           fpu vme de pse tsc msr pae mce cx8 apic se
                                 p mtrr pge mca cmov pat pse36 clflush mmx 
                                 fxsr sse sse2 ss ht syscall nx pdpe1gb rdt
                                 scp lm constant_tsc rep_good nopl xtopolog
                                 y nonstop_tsc cpuid tsc_known_freq pni pcl
                                 mulqdq vmx ssse3 fma cx16 pcid sse4_1 sse4
                                 _2 x2apic movbe popcnt aes xsave avx f16c 
                                 rdrand hypervisor lahf_lm abm 3dnowprefetc
                                 h invpcid_single ssbd ibrs ibpb stibp ibrs
                                 _enhanced tpr_shadow vnmi flexpriority ept
                                  vpid fsgsbase tsc_adjust bmi1 hle avx2 sm
                                 ep bmi2 erms invpcid rtm avx512f avx512dq 
                                 rdseed adx smap clflushopt clwb avx512cd a
                                 vx512bw avx512vl xsaveopt xsavec xgetbv1 x
                                 saves arat avx512_vnni md_clear arch_capab
                                 ilities

---------------Memory Info-----------------------
              total        used        free      shared  buff/cache   available
Mem:           15Gi       936Mi        12Gi       5.0Mi       2.1Gi        14Gi
Swap:            0B          0B          0B

---------------Nic Info--------------------------
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq master pnet0 state UP mode DEFAULT group default qlen 1000

---------------IP Info---------------------------
●   State: routable                          

---------------Bridge Info-----------------------
pnet0           8000.42010a9a0022       no              eth0
pnet1           8000.000000000000       no
pnet2           8000.000000000000       no
pnet3           8000.000000000000       no
pnet4           8000.000000000000       no
pnet5           8000.000000000000       no
pnet6           8000.000000000000       no
pnet7           8000.000000000000       no
pnet8           8000.000000000000       no
pnet9           8000.000000000000       no

---------------H/W Accel-------------------------
INFO: /dev/kvm exists
KVM acceleration can be used

---------------Service Info----------------------
-------------------------------------------------
--------------Guacamole--------------------------
● guacd.service - LSB: Guacamole proxy daemon
     Loaded: loaded (/etc/init.d/guacd; generated)
     Active: active (running) since Thu 2022-12-08 09:04:34 UTC; 19min ago

--------------Tomcat-----------------------------
Unit tomcat8.service could not be found.

--------------Mysql------------------------------
● mysql.service - MySQL Community Server
     Loaded: loaded (/lib/systemd/system/mysql.service; enabled; vendor preset: enabled)
     Active: active (running) since Thu 2022-12-08 09:04:36 UTC; 19min ago

--------------Apache-----------------------------
● apache2.service - The Apache HTTP Server
     Loaded: loaded (/lib/systemd/system/apache2.service; enabled; vendor preset: enabled)
     Active: active (running) since Thu 2022-12-08 09:04:35 UTC; 19min ago

root@eve-conn-v5:~# 
You do not have the required permissions to view the files attached to this post.

ymeillier
Posts: 10
Joined: Mon Feb 03, 2020 1:30 am

Re: eve-comm on GCP: kernel crash from uksm

Post by ymeillier » Fri Dec 09, 2022 3:58 am

No idea how you created EVE community on GCP
I didn't 'create eve comm on GCP'. I'm running eve following the GCP installation instructions of the eve-comm cookbook.

Used EVE-Community cookbook to deploy it.
same here.

I found the exact steps that trigger the issues. The trace could be a red herring although the trace consistently surfaces every time the issue occurs.

If you want to reproduce the issue do this:
  • I don't think the image matters, i tried with vyatta vyos nodes and unbuntu 20.004 nodes as well.
    Add 8 nodes and connect them all to pnet0/cloud0
    Start all nodes at once or 1 after the other. If don't hit the issue at first power off and try again.
    you will lose connectivity to the gui and won't be able to ssh or even ping the GCE instance hosting eve.
If i repeat the test (after instance reboot) with the nodes connected to cloud1, the issue never occurs.
The network stack gets screwed up and a reboot of the instance is required.

Repeat the above steps and you will be able to observe the same problem as i do.

Note that all of the above is done on brand new GCE instance and eve installed following the comm cookbook verbatum.

ymeillier
Posts: 10
Joined: Mon Feb 03, 2020 1:30 am

Re: eve-comm on GCP: kernel crash from uksm

Post by ymeillier » Fri Dec 09, 2022 4:02 am

also, in the lab topo you provided you have your nodes connected to each other. That works for me too.
You need to connect the nodes to cloud0 ( i use that to ssh to the nodes to configure them so neend pnet0 bridge).
Craete 8 vyos nodes all connected to cloud0. power them on. youll experience the crash. if not at first try again. even if just one by one... at some point the nodes won't power on, and you will connectivity to the GUI/server

Uldis (UD)
Posts: 5067
Joined: Wed Mar 15, 2017 4:44 pm
Location: London
Contact:

Re: eve-comm on GCP: kernel crash from uksm

Post by Uldis (UD) » Fri Dec 09, 2022 6:37 pm

Cloud0 is not is use for GCP, it has only single IP and reason to connect your routers to it

Post Reply