[2021-07-02T16:34:10.283Z] WARN: cmon-agent/6223 on cn01: collectors-common/plugin/jvm/11db239b-3c76-445a-b5f2-0321c8b0b78c metrics unavailable, treating as empty
NotAvailableError: jvm: failed to execute: exec "/opt/custom/cmon/vm-plugins/jvm.prom 11db239b-3c76-445a-b5f2-0321c8b0b78c": unexpectedly terminated by signal SIGKILL
at _afterExecution (/opt/smartdc/agents/lib/node_modules/cmon-agent/lib/instrumenter/collectors-common/plugin.js:221:34)
at _onForkWaited (/opt/smartdc/agents/lib/node_modules/cmon-agent/lib/plugin-executor.js:60:9)
at onChildExited (/opt/smartdc/agents/lib/node_modules/cmon-agent/node_modules/forkexec/lib/forkexec.js:83:2)
at /opt/smartdc/agents/lib/node_modules/cmon-agent/node_modules/forkexec/lib/forkexec.js:53:4
at ChildProcess.exithandler (child_process.js:207:5)
at emitTwo (events.js:87:13)
at ChildProcess.emit (events.js:172:7)
function is_docker | |
{ | |
if [ "${container}" = "docker" ]; then | |
return 0 | |
else | |
return 1 | |
fi | |
} | |
function is_triton |
version: '3' | |
services: | |
nginx: | |
image: nginx | |
container_name: nginx | |
restart: always | |
ports: | |
- "80:80" | |
- "443:443" | |
volumes: |
Every once in a while I will encounter the following error on a VM reprovision:
Command failed: imgadm get: error (ImageNotInstalled): image "372d550b-b224-46ec-950b-0c6a2421cad4"
was not found on zpool "undefined"
I added logging to the reprovision()
function in VM.js
which logs vmobj
right after it's populated from this line:
vmobj = vminfodEventGetVM(uuid, ready_ev);
Oracle 12 forms can’t run on SmartOS LX zone
Oracle Forms 12 requires the following Linux kernel parameters to be set but the SmartOS LX emulation does not map these kernel parameters to Illumos kernel parameters.
sysctl -w net.core.rmem_max=2096304
sysctl -w net.core.wmem_max=2096304
We were able to set the equivalent underlying Illumos kernel parameters with the command below but the issue is that the LX emulation lies to the Linux system call and returns some arbitrary value even though we have changed the kernel param in Illumos under the hood.
While the CN is still running in the source Triton DC, follow the steps in https://docs.joyent.com/private-cloud/troubleshooting/reset but skip the factory reset step because that is what tells SmartOS to destroy the zpool. Once the CN has been forgotten from the source Triton DC then power it down, cable it up to the target DC and when it iPXE boots the first time Triton will see it as already setup so it won’t give you the option to set it up. Replace the DC name and necessary IP addresses in the following files on the CN:
/opt/smartdc/agents/etc/vm-agent.config.json
/opt/smartdc/agents/etc/net-agent.config.json
/opt/smartdc/agents/etc/cn-agent.config.json
/opt/smartdc/registrar/etc/config.json
/opt/smartdc/config/node.config
/opt/smartdc/agents/etc/smartlogin.cfg
[root@cn02 (us-west-agc) ~]# zfs get mounted,mountpoint zones/cores/a2d5015e-e073-4758-b37f-44aa2aabddbf
NAME PROPERTY VALUE SOURCE
zones/cores/a2d5015e-e073-4758-b37f-44aa2aabddbf mounted yes -
zones/cores/a2d5015e-e073-4758-b37f-44aa2aabddbf mountpoint /zones/a2d5015e-e073-4758-b37f-44aa2aabddbf/cores local
[root@cn02 (us-west-agc) ~]# zfs umount /zones/a2d5015e-e073-4758-b37f-44aa2aabddbf/cores
cannot unmount '/zones/a2d5015e-e073-4758-b37f-44aa2aabddbf/cores': not a mountpoint
[root@headnode (us-west-rno) ~]# sdc-server list
HOSTNAME UUID VERSION SETUP STATUS RAM ADMIN_IP
cn01 44454c4c-3300-1050-8051-b5c04f393033 7.0 true unknown 1048193 10.10.65.50
headnode 44454c4c-4c00-1039-8053-b9c04f4a4832 7.0 true running 262049 10.10.65.10
[root@headnode (us-west-rno) ~]# sdc-oneachnode -n headnode uptime
{"name":"sdc-oneachnode","hostname":"headnode","pid":215433,"component":"URConnection","level":50,"err":{"message":"connect timeout","name":"Error","stack":"Error: connect timeout\n at null.<anonymous> (/zones/690741eb-28bf-4920-9eac-bcbc728cdf4f/root/opt/smartdc/sdc/node_modules/urclient/lib/mq.js:178:13)\n at Timer.listOnTimeout [as ontimeout] (timers.js:121:15)"},"msg":"amqp error","time":"2021-05-14T15:10:11.012Z","v":0}
VError: amqp error: connect timeout
at URConnection._disconnect (/zones/690741eb-28bf-4920-9eac-bcbc728cdf4f/root/opt/smartdc/sdc/n
[root@e4-43-4b-9c-b0-a9 ~]# tail -20 $(svcs -L filesystem/smartdc)
+ is_encr=aes-256-gcm
+ [[ aes-256-gcm != \o\f\f ]]
+ unlock_pool zones
+ local pool=zones
++ zfs get -Hpo value keystatus zones
+ local keystatus=unavailable
+ [[ unavailable == \a\v\a\i\l\a\b\l\e ]]
+ kbmadm unlock zones
[root@9676e574-47e5-4987-b116-e712e5719e41 (us-west-agc:kbmapi0) ~]# kbmctl recovery ls
UUID STAGED_TOKENS ACTIVATED_TOKENS STATE
76f89dae-0399-5456-bf64-8d1574033076 0 2 active
00da4496-a19c-5aa0-acaa-599f51f5f20f 0 0 staged