Initial definition of a Transaction Processing System (TPS)#1898
Initial definition of a Transaction Processing System (TPS)#1898rrschulze wants to merge 32 commits intoopen-telemetry:mainfrom
Conversation
|
This PR was marked stale due to lack of activity. It will be closed in 7 days. |
|
Closed as inactive. Feel free to reopen if this PR is still being worked on. |
mtwo
left a comment
There was a problem hiding this comment.
We created / reviewed this within the mainframe SIG prior to submitting it here. Approving!
gshriver
left a comment
There was a problem hiding this comment.
Looks good to me, and matches the discussions in SIG group to this point.
docs/attributes-registry/ibm.md
Outdated
| <!-- NOTE: THIS FILE IS AUTOGENERATED. DO NOT EDIT BY HAND. --> | ||
| <!-- see templates/registry/markdown/attribute_namespace.md.j2 --> | ||
|
|
||
| # Ibm |
There was a problem hiding this comment.
You probably want IBM here right? You can add the acronym in this file and then it should be all uppercase: https://github.com/open-telemetry/semantic-conventions/blob/main/templates/registry/markdown/weaver.yaml. Also for the other acronyms.
|
This PR was marked stale due to lack of activity. It will be closed in 7 days. |
|
This PR was marked stale due to lack of activity. It will be closed in 7 days. |
|
Closed as inactive. Feel free to reopen if this PR is still being worked on. |
model/tps/registry.yaml
Outdated
| brief: > | ||
| This group defines generic attributes for Transaction Processing Systems (TPS). | ||
| attributes: | ||
| - id: tps.system |
There was a problem hiding this comment.
should it be tps.name? TPS already has system in it.
There was a problem hiding this comment.
Looking at the current spec of db and messaging, it feels inconsistent to skip 'system'. How about going with tps.system.name?
model/tps/registry.yaml
Outdated
| - id: tps.region.id | ||
| type: string | ||
| stability: experimental | ||
| brief: "Runtime environment within the Transaction Processing System" |
There was a problem hiding this comment.
this does not seem to match attribute name
model/tps/registry.yaml
Outdated
| type: string | ||
| stability: experimental | ||
| brief: Determines the type of commit mode processing that IMS performs. | ||
| examples: ['0', '1'] |
model/tps/spans.yaml
Outdated
| requirement_level: required | ||
| - ref: tps.transaction.id | ||
| requirement_level: required | ||
| - ref: tps.region.id |
There was a problem hiding this comment.
should it be a resource attribute instead?
There was a problem hiding this comment.
@lmolkova you're right about that. This needs to be identifying attribute of a tps.region entity (on z/OS). Let me take this back to the mainframe SIG.
There was a problem hiding this comment.
Moving tps.region.id into a tps entity for the transaction processing system.
|
This PR was marked stale due to lack of activity. It will be closed in 7 days. |
…inition of the TPS server span
Co-authored-by: Liudmila Molkova <neskazu@gmail.com>
…e/semantic-conventions into rrschulze/issue1687_tps Update with suggestions on PR
|
This PR has been labeled as stale due to lack of activity. It will be automatically closed if there is no further activity over the next 7 days. |
|
Will discuss PR in today's Semantic Conventions SIG call. |
| stability: development | ||
| brief: 'This document defines semantic conventions for a server spans of Transaction Processing Systems' | ||
| note: | | ||
| **Span name** SHOULD be `{tps.system.name} {tps.transaction.id}`. |
There was a problem hiding this comment.
ptal at the span guidance - https://github.com/open-telemetry/semantic-conventions/blob/main/docs/how-to-write-conventions/README.md#naming-pattern - this is best practice, low-cardinality of a span name is arequirement
| - ref: tps.program.name | ||
| - ref: tps.unit_of_work.id | ||
| - ref: tps.connection.name | ||
| - ref: tps.facility.type |
|
|
||
| **Status:**  | ||
|
|
||
| This document defines semantic conventions for a server spans of Transaction Processing Systems |
There was a problem hiding this comment.
maybe polish this to describe what this span represents
There was a problem hiding this comment.
SIG discussion:
- IBM and Oracle are two systems that would implement this
- Oracle is not involved at the moment
- If it's unlikely to get a second system to support this (or review and bless), it's better to do ibm or mainframe-specific convention
There was a problem hiding this comment.
@lmolkova we had a discussion in the Mainframe SIG. Having an ibm namespace is probably not a good idea.
- there are multiple operating systems and servers sold by IBM that could create some confusion
- There are multiple vendors that provide middleware and operating systems that run on the IBM servers
- There is a lot of synergy between the mainframe capabilities, for example virtualization, database, transaction processing, messaging, with other server architectures
As a result, the conclusion of the sig is that the high level qualifier should be closer to the z server technology than to ibm itself.
| - id: entity.tps | ||
| type: entity | ||
| stability: development | ||
| name: tps |
There was a problem hiding this comment.
nit, this would be more precise an future-proof
| name: tps | |
| name: tps.region |
|
This PR has been labeled as stale due to lack of activity. It will be automatically closed if there is no further activity over the next 7 days. |
Fixes #1687
Changes
The PR is based on the input from the Mainframe SIG meetings and is intended to have the discussion to complete the definition of a generic Transaction Processing System. Basing on the generic TPS, we aim to finalize the span definitions for CICS and IMS.
The PR adds the following new components to the semantic conventions:
tps: Introduces new (generic) Transaction Processing System (TPS), an initial set of attributes and defines TPS spans extending the HTTP and RPC servers spans, and uses them for initial definition of server spans for the TPS CICS and IMS available on the mainframe.
ibm: Introduces ibm as name space to represent specific attributes of systems like ibm.cics and ibm.ims (aiming to align with guidance for system names described in #1708
Note: if the PR is touching an area that is not listed in the existing areas, or the area does not have sufficient domain experts coverage, the PR might be tagged as experts needed and move slowly until experts are identified.
Merge requirement checklist
[chore]