aboutsummaryrefslogtreecommitdiffstats
path: root/src/mtp_layer3.c
AgeCommit message (Collapse)AuthorFilesLines
2013-03-19link_sets: Move the submit for SCCP/ISUP into a function pointerzecke/multiple-link-set-typesHolger Hans Peter Freyther1-1/+6
In the preparation of supporting multiple link_sets types the send routines are accessed through function pointers now.
2013-03-19link_sets: Remove direct function calls for data/up/down from MTP3Holger Hans Peter Freyther1-1/+2
We want to support different link_sets types. To do this we will need to reduce the coupling of mtp_link_set with the application. This is the first step to remove the notifications. This comes with a performance cost but it has not been measured. In the long run the conditions for if (set->on_bla) should be removed as this indicates a wrong configuration of the linkset.
2012-01-16mtp: Allow to send SCCP/ISUP to a specific endpointHolger Hans Peter Freyther1-2/+7
For a linkset define where SCCP/ISUP should be send. This config should probably move up to the application part when real work on the routing is done. Right now the sccp_opc/sccp_dpc need to stay inside the mtp_layer3.c to be able to send a TFA for the reachable OPC and it is easier to keep both (dpc/opc) in the same file.
2012-01-16mtp: Provide the method with the OPC and DPCHolger Hans Peter Freyther1-6/+7
We want to submit on a link set but want to provide the OPC and DPC of the callers choice.
2012-01-16mtp: Name the parameter opc instead of just PCHolger Hans Peter Freyther1-3/+3
2011-05-08Adopt to recent libosmocore namespace changes and libosmogsmHarald Welte1-10/+10
2011-03-03ss7: Drop the input of packages as wellzecke/test-forward-failureHolger Hans Peter Freyther1-1/+1
We don't want the input change any state on the linkset and will drop them if we think our application is not reachable.
2011-03-03ss7: Move the blocking of outgoing messages into the SS7 appHolger Hans Peter Freyther1-18/+0
We are using knowledge of the SS7 application to drop outgoing packages to force failures on the link and should move this into the ss7 application.
2011-03-03ss7: Do not send anything until both linksets in an app are upHolger Hans Peter Freyther1-0/+10
We need some way to forward the failure of one link to another but they are not normally routed so we can not send a TFP. Right now we will simply stop responding until both links are up. This should make the SLTM fail and trigger a re-alignment on both sides. The key here is that the 2 * SLTM timeout needs to be higher than it takes to re-align the link. I'm not sure this code will work.
2011-03-02mtp: Consider the linkset be ready for SCTP after having collected routing dataHolger Hans Peter Freyther1-1/+1
After the expiry of T18 we should have collected the routing data from the adjacent links and should be able to send SCCP packages to remote endpoints.
2011-02-23mtp: Call the mtp_link_set always set otherwise things get confusingHolger Hans Peter Freyther1-78/+78
In a backtrace it is confusing to see variables called link and link and one is a mtp_link and the other is a mtp_link_set.
2011-02-23mtp: Attempt to implement the MTP Restart properlyHolger Hans Peter Freyther1-20/+75
For the SSP functionatilty we will need to have the timers T18 and T20. In the period of T18 we will collect TFP/TFR/TFA for the reachable nodes of the system. Each of this node will send us a TRA when it is finished. Right now we assume to only have one node and stop the T18 after the TRA of this node. Then we would need to send the TFP/TFR we have collected. On the expiry of the T20 timer we will need to send our TRA and notify local users. For more complex routing we will need to have a shared routing cache and remember which SSNs and OPCs are reachable and have inter linkset notifications.
2011-02-23vty: Add all mighty new vty interface for osmo-stpHolger Hans Peter Freyther1-0/+3
This new interface allows to have multiple linksets, msc connections and ways to connect those in one instance of the osmo-stp. Forbid to reset linksets without an app.
2011-02-22misc: Use llist_add_tail to add new links to the endHolger Hans Peter Freyther1-1/+1
2011-02-17log: Improve some logging messages across several files.Holger Hans Peter Freyther1-15/+20
2011-02-17mtp: Rename link_no to nr, add a name like for the other commandsHolger Hans Peter Freyther1-2/+3
2011-02-17mtp: Make the mtp_link point to a specific type of linkHolger Hans Peter Freyther1-12/+0
We might want to be able to change the type of a link at runtime. Decouple the link and the actual type of the link.
2011-02-17mtp: Stop hardcoding the supported SSN inside the mtp_layer3.cHolger Hans Peter Freyther1-2/+1
Stop hardcoding the supported ssn's inside the mtp_layer3.c and make it possible to allow to configure this in the future.
2011-02-17mtp: Rename no to nr to align with the rest of the structsHolger Hans Peter Freyther1-2/+2
2011-02-17mtp: Allow to find a linkset by numberHolger Hans Peter Freyther1-26/+37
2011-02-17mtp: Allocate the MTPLinkSet as a child of the BSCHolger Hans Peter Freyther1-11/+6
In preparation of the VTY code change, make the mtp linkset a child of the bsc.
2011-02-10stp: Add code that allows to inject messages on any linksetHolger Hans Peter Freyther1-2/+19
This can be useful to test out certain messages without having any of the linksets be fully connected. It is not possible to get the result. In the future this code should reply with an M2UA error message if something went wrong.
2011-01-28mtp: Turn error messages into full sentences.Holger Hans Peter Freyther1-2/+2
Turn the error messages into full sentences that end with a full stop. Try to put the link set name and the link number into these messages.
2011-01-26mtp: Assign a name to linkset, use this name in the log messagesHolger Hans Peter Freyther1-4/+5
2011-01-25mtp: Send the TFP/TRA/TFA with the first SLS provided by that link1.1.0Holger Hans Peter Freyther1-4/+5
2011-01-25mtp: Make SCCP tfa/tra/tfp messages work on the linksetHolger Hans Peter Freyther1-22/+24
This will be used to send the data with the right SLS from the right link of the linkset.
2011-01-25mtp: Remember the first SLS this link is onHolger Hans Peter Freyther1-1/+7
2011-01-24stats: Fix counting incoming packets from a linkHolger Hans Peter Freyther1-0/+1
2011-01-24mtp: Call it mtp_link_submit and use it in mtp_linkHolger Hans Peter Freyther1-7/+7
Remove the _set from the API, call it from the mtp_link.c. This will fix the statistics for outgoing packets.
2011-01-22stats: Count packets on a per link levelHolger Hans Peter Freyther1-2/+5
2011-01-22stats: Collect statistics on the linksetHolger Hans Peter Freyther1-0/+24
Collect incoming, in sccp, in isup, out isup, out sccp packets. This does not yet work for out in total.
2011-01-20mtp: Return SLTA on the same SLS, send SLTM with a static slsHolger Hans Peter Freyther1-2/+7
Derive the SLS from the static link_no we have assigned.
2011-01-20mtp: Send the SLTM/SLTA on the link directlyHolger Hans Peter Freyther1-6/+1
For both the SLTM that the app is sending and the SLTM we are receiving we don't want to go through the SLC.
2011-01-20mtp: Count the links in a linksetHolger Hans Peter Freyther1-0/+1
2011-01-20mtp: Remove the unused linkset_sccp_down routine as it is not usedHolger Hans Peter Freyther1-4/+0
2011-01-20mtp: Move the link testing into a new file for mtp_linkHolger Hans Peter Freyther1-123/+28
The link testing should work under the linkset and if the SLTM is failing the link should be taken down and it should be restarted.
2011-01-20mtp: Rename the the_link to set as it is a link setHolger Hans Peter Freyther1-5/+6
2011-01-20mtp: Start the linktest directlyHolger Hans Peter Freyther1-9/+1
No need for delaying the linktest, start it directly when the linkset should come up
2011-01-20mtp: Send the TFP/TRA once the first link is availableHolger Hans Peter Freyther1-21/+39
Instead of sending SCCP TFP/TRA as a response to a TRA the code will now send one as a response to the first SLTA we receive in the linkset on the first link.
2011-01-20mtp: Only respond to regular maintainenance for our code pointHolger Hans Peter Freyther1-0/+6
2011-01-20isup: Allow to have a different OPC for ISUP messages.Holger Hans Peter Freyther1-1/+1
2011-01-20mtp: Send a TFA for the point codes we are routing to.Holger Hans Peter Freyther1-4/+24
2011-01-20mtp: Send TRA messages on all affected point codesHolger Hans Peter Freyther1-5/+20
2011-01-20mtp: Send the TFP for every POC we want to handle.Holger Hans Peter Freyther1-3/+20
2011-01-20mtp: Make the input path work on a specific link of the linksetHolger Hans Peter Freyther1-6/+6
2011-01-20stp: Forward unhandled ISUP from one end to anotherHolger Hans Peter Freyther1-1/+1
2011-01-20mtp: Do not send anything in case the there is no linkHolger Hans Peter Freyther1-0/+3
If there is no usable link in the linkset we should not try to send. This assumes that all incoming packages come from links that are up. We do not check this right now but we will have to check this.
2011-01-20mtp: Send the SSA/SSP for the apoc and assn of the requestHolger Hans Peter Freyther1-3/+3
2011-01-20mtp: Send the SSA with the affected POC for SCCP.Holger Hans Peter Freyther1-1/+1
2011-01-17mtp: Rename link_data to mtp_link and move out the transport specific thingsHolger Hans Peter Freyther1-5/+5
Rename link_data to mtp_link and move it into the mtp_data header file, also remove the union to ease creating more of the subtypes. This is done in preparation to the linkset knowing more about the link (e.g. having a link test per link instead of per link).