[202511][cherry][hft]: Remove IPFIX template size estimation, query actual size from …#4369
Merged
vmittal-msft merged 1 commit intosonic-net:202511from Mar 19, 2026
Conversation
…SAI instead (sonic-net#4304) What I did Removed the IPFIX template buffer size estimation logic in HFTelProfile::updateTemplates() and replaced it with the standard SAI two-phase query pattern: first query the required size with count=0 / list=nullptr, then allocate and fetch. Why I did it The previous estimation logic used hardcoded constants to predict the IPFIX template buffer size, but could underestimate (e.g., estimated 65535 bytes vs. actual 119352 bytes). This caused the SDK to log ERR-level messages on the first attempt: ERR syncd#SDK: [SAI_TAM.ERR] mlnx_generate_ipfix_templates: Buffer size is too small to hold IPFIX template [size:65535, required:119352]. ERR syncd#SDK: [SAI_TAM.ERR] mlnx_tam_tel_type_get_ipfix_templates: Failed to generate IPFIX templates. ERR syncd#SDK: [SAI_TAM.ERR] mlnx_tam_tel_type_attrib_get: Failed to get attribute. ERR syncd#SDK: [SAI_UTILS.ERR] get_dispatch_attribs_handler: Failed Get #0, IPFIX_TEMPLATES, key:TAM_TEL_TYPE Although the retry with SAI_STATUS_BUFFER_OVERFLOW worked correctly, these error logs caused test_hft_full_queue_counters to fail in LogAnalyzer teardown. By always querying the size first (count=0, list=nullptr), we avoid the unreliable estimation entirely and follow the idiomatic SAI pattern for variable-length attributes. Signed-off-by: Priyansh Tratiya <[email protected]>
Collaborator
|
/azp run |
|
Azure Pipelines successfully started running 1 pipeline(s). |
vmittal-msft
approved these changes
Mar 19, 2026
1 task
andrewsapronov
pushed a commit
to netbergtw/sonic-swss
that referenced
this pull request
Mar 25, 2026
…SAI instead (sonic-net#4304) (sonic-net#4369) What I did Removed the IPFIX template buffer size estimation logic in HFTelProfile::updateTemplates() and replaced it with the standard SAI two-phase query pattern: first query the required size with count=0 / list=nullptr, then allocate and fetch. Why I did it The previous estimation logic used hardcoded constants to predict the IPFIX template buffer size, but could underestimate (e.g., estimated 65535 bytes vs. actual 119352 bytes). This caused the SDK to log ERR-level messages on the first attempt: ERR syncd#SDK: [SAI_TAM.ERR] mlnx_generate_ipfix_templates: Buffer size is too small to hold IPFIX template [size:65535, required:119352]. ERR syncd#SDK: [SAI_TAM.ERR] mlnx_tam_tel_type_get_ipfix_templates: Failed to generate IPFIX templates. ERR syncd#SDK: [SAI_TAM.ERR] mlnx_tam_tel_type_attrib_get: Failed to get attribute. ERR syncd#SDK: [SAI_UTILS.ERR] get_dispatch_attribs_handler: Failed Get #0, IPFIX_TEMPLATES, key:TAM_TEL_TYPE Although the retry with SAI_STATUS_BUFFER_OVERFLOW worked correctly, these error logs caused test_hft_full_queue_counters to fail in LogAnalyzer teardown. By always querying the size first (count=0, list=nullptr), we avoid the unreliable estimation entirely and follow the idiomatic SAI pattern for variable-length attributes. Signed-off-by: Priyansh Tratiya <[email protected]> Co-authored-by: Ze Gan <[email protected]>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Cherry-pick: #4304
…SAI instead (#4304)
What I did
Removed the IPFIX template buffer size estimation logic in HFTelProfile::updateTemplates() and replaced it with the standard SAI two-phase query pattern: first query the required size with count=0 / list=nullptr, then allocate and fetch.
Why I did it
The previous estimation logic used hardcoded constants to predict the IPFIX template buffer size, but could underestimate (e.g., estimated 65535 bytes vs. actual 119352 bytes). This caused the SDK to log ERR-level messages on the first attempt:
ERR syncd#SDK: [SAI_TAM.ERR] mlnx_generate_ipfix_templates: Buffer size is too small to hold IPFIX template [size:65535, required:119352]. ERR syncd#SDK: [SAI_TAM.ERR] mlnx_tam_tel_type_get_ipfix_templates: Failed to generate IPFIX templates. ERR syncd#SDK: [SAI_TAM.ERR] mlnx_tam_tel_type_attrib_get: Failed to get attribute. ERR syncd#SDK: [SAI_UTILS.ERR] get_dispatch_attribs_handler: Failed Get #0, IPFIX_TEMPLATES, key:TAM_TEL_TYPE Although the retry with SAI_STATUS_BUFFER_OVERFLOW worked correctly, these error logs caused test_hft_full_queue_counters to fail in LogAnalyzer teardown.
By always querying the size first (count=0, list=nullptr), we avoid the unreliable estimation entirely and follow the idiomatic SAI pattern for variable-length attributes.
What I did
Why I did it
How I verified it
Details if related