Attention is currently required from: laforge.

View Change

1 comment:

• Patchset:

  • Patch Set #1:

    I think I'd prefer something like @if (nokia_bts_type(bts) == NOKIA_BTS_INSITE)@ as it would scale m […]

    Right now we have the following situation:

    • Out of supported and tested models, MetroSite and UltraSite require one handling (currently treated as default), InSite requires different handling (was there from the beginning) and Flexi MR requires third handling, added with my third patch in this series.
    • After Flexi MR, there is Multiradio 10, the last model to support E1 Abis via add-on card. I don't have an MR10 system unit yet, but I see plenty of them on ebay, and I plan on buying one for testing. It works with the same radio modules as plain Flexi MR. I expect MR10 to require the same handling in osmo-bsc as plain FMR, and my third patch in this series currently assumes so.
    • In between UltraSite and Flexi MR there is also Flexi EDGE, intermediate evolutionary step. However, I and A2GC are essentially skipping Flexi EDGE, going directly to Flexi MR or MR10. It is not clear whether or not anyone else will ever invest time and money into acquiring Flexi EDGE hw and putting it together just to test osmo-bsc on it - hence it may remain forever in unknown state.
    • After MR10, all newer models are IP Abis only. Csaba and I are looking into the possibility of implementing IP Abis support on Flexi MR and MR10, using hw in my lab, as a stepping stone toward subsequent IP-only models - but nothing is certain with this idea.

    Going back to your idea of a `nokia_bts_type(bts)` function that distills Nokia's own BTS model ID codes into a smaller set of `NOKIA_BTS_*` return values that capture different handling required in osmo-bsc, right now I would feel a little uneasy defining and implementing such grouping: we don't know if Flexi EDGE should be handled like Metro/UltraSite or like Flexi MR, and the use of exactly the same handling between MR and MR10 remains to be tested on the latter.

    Since you said this issue is not a blocker, if I don't see any further comments in the next couple of days, I'll CR+2 to myself (based on CR+1 from you and @pespin@sysmocom.de) and merge.

To view, visit change 42725. To unsubscribe, or for help writing mail filters, visit settings.