 Adam Ierymenko
|
29ec7bf3a2
Add more specific check in source==self case instead of dumping it.
|
8 years ago |
|
Adam Ierymenko
|
60ff280dcb
Another tweak to cluster I/O rules.
|
8 years ago |
|
Adam Ierymenko
|
b378f5dcd7
Take 3
|
8 years ago |
|
Adam Ierymenko
|
5e11cf6378
Can't armor() a packet until all flags are set.
|
8 years ago |
|
Adam Ierymenko
|
5dbebc513a
Minor send path refactor to make packet I/O work on clusters if they are members of networks. Also fix a crash if compiled in cluster mode but no cluster is enabled.
|
8 years ago |
|
Adam Ierymenko
|
eebd271bb1
Implement cross cluster sharing of network configs to make clusters able to actually join networks.
|
8 years ago |
|
Adam Ierymenko
|
a963810e02
Tweak a few cluster params for higher perf / lower initial setup latency.
|
9 years ago |
|
Adam Ierymenko
|
4e4fd51117
boring doc stuff
|
9 years ago |
|
Adam Ierymenko
|
451b8aa7b2
Query both root and network controller for multicast last resort GATHER.
|
9 years ago |
|
Adam Ierymenko
|
2854f14966
Some cluster parameter tweaks, and change the test code in tests/http to dramatically reduce the amount of data transfer to/from the test master. Also add results of first successful 50k test.
|
9 years ago |
|
Adam Ierymenko
|
b171c9a0db
Replace slow std::list<> with an O(log(N)) data structure for the cluster relaying send queue.
|
9 years ago |
|
Adam Ierymenko
|
32ec378e3b
Announce that we have peers on the cluster when we first see them to improve startup times, and add a result crunching script to tests/http.
|
9 years ago |
|
Adam Ierymenko
|
35c4e28f31
Mark geo-redirected paths as suboptimal and do not report that we have a peer if all we have is one of these. Also a few other small fixes.
|
9 years ago |
|
Adam Ierymenko
|
73e2c6e511
How did that ever work?
|
9 years ago |
|
Adam Ierymenko
|
4ad2ff2f71
.
|
9 years ago |
|
Adam Ierymenko
|
b57c855a8c
PROXY_SEND fix.
|
9 years ago |
|
Adam Ierymenko
|
ff81415ffe
Build fixes...
|
9 years ago |
|
Adam Ierymenko
|
57b71bfff0
Cluster simplification and refactor work in progress...
|
9 years ago |
|
Adam Ierymenko
|
6bc8c9d8ef
Clustering cleanup, still a work in progress.
|
9 years ago |
|
Adam Ierymenko
|
5f39d5b7ea
Further pare down Cluster messaging and rename some stuff.
|
9 years ago |
|
Adam Ierymenko
|
8ea5f424fd
Small amount of cleanup.
|
9 years ago |
|
Adam Ierymenko
|
ba5008c370
Now make cluster announcements slightly more frequent.
|
9 years ago |
|
Adam Ierymenko
|
3fa1b5a89d
Make cluster HAVE_PEER a little lighter by removing full identity. Technically this is not necessary and they will propagate over time in any case. Might restore it in some form in the future.
|
9 years ago |
|
Adam Ierymenko
|
7fbe2f7adf
Tweak some more timings for better reliability.
|
9 years ago |
|
Adam Ierymenko
|
60ce886605
Tweak some timings for better reliability.
|
9 years ago |
|
Adam Ierymenko
|
d6c0d176ee
Periodically re-announce peers that we have.
|
9 years ago |
|
Adam Ierymenko
|
883c84bdb9
Tweak some timings, and remove some dead code.
|
9 years ago |
|
Adam Ierymenko
|
51fcc75354
Some cleanup, and use getPeerNoCache() exclusively in Cluster.
|
9 years ago |
|
Adam Ierymenko
|
cc1b275ad9
Replicate peer endpoints and forget paths if we have them -- this allows two clusters to talk to each other, whereas forgetting all paths does not.
|
9 years ago |
|
Adam Ierymenko
|
cc6080fe38
(1) No need to confirm if we are a root (small optimization), (2) Refactor peer affinity tracking.
|
9 years ago |
