Hyposcaler bot/issue 14

[hyposcaler-bot]

Updates lab to use stages with wait-for
Corrected router configs to adjust for changes in BGP policy since lab was created
Pinned srlinux nodes to 25.3.2

This should address issue #14

Note: this was only lightly tested, I only confirmed that the lab starts without error

[hellt]

@azyablov it seems you were outpaced =) but would be good for you to give it a go

[hellt]

thanks @hyposcaler-bot

[azyablov]

@hellt in my to do list ) will check today

[azyablov]

@hellt process.name and hostname keywords got disappeared with new SRL version, investigating. Looking for ECS 9.0.0 and updated dashboards as well.

[hellt]

I think the big change that has happened in the SRL land is the RFC5424 SYSLOG support; at the time of this post we did not have it

[azyablov]

@hyposcaler-bot @hellt please check accumulated changes adjusted-issue-14

1. Reworked transformation pipeline for Logstash.
2. Switched to ECS v8 with automatic template provisioning.
3. Added provisioning client to avoid manual steps: health checks, password updates, dashboards and data view.
4. Activated TLS and Auth, since plain HTTP to be deprecated.
5. Rewritten shell scripts in more idiomatic way.
6. Docs structure now follows ECSv8.
7. Added syslog-ng for verification, trouble shooting and SIEM cases. Script to provision on SRL NEs.

Potential issues found:
syslog inputs plugin can't coupe with RFC5424 correctly, while SRL send compliant messages

"tags" => [
        [0] "syslog",
        [1] "srlinux",
        [2] "_grokparsefailure_sysloginput"
    ],

Opened issue logstash-input-syslog/issues/79

[azyablov]

Regrading pull request:

• NEs configs are ok.
• clab file logic has been changed in adjusted-issue-14 order to start SRL nodes after provisioner, which allows to capture logs immediately after the start having ELK stack ready for service before.