LBRYCommunity/lbrycrd
1
0
Fork 0
Code Issues 29 Pull requests 1 Projects Releases 32 Packages Wiki Activity

doc: add comment explaining initial header request

Browse source 
Add a comment that explains why the initial "getheader" requests are
made starting from the block preceding the currently best one.

Thanks to sdaftuar for the explanation!
This commit is contained in:
Daniel Kraft 2015-10-14 20:42:49 +02:00
parent a1d623da3e
commit 53b86d0de5
1 changed files with 10 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

11
src/main.cpp
View file

@ -4955,7 +4955,16 @@ bool SendMessages(CNode* pto, bool fSendTrickle)
if ((nSyncStarted == 0 && fFetch) || pindexBestHeader->GetBlockTime() > GetAdjustedTime() - 24 * 60 * 60) {
state.fSyncStarted = true;
nSyncStarted++;
CBlockIndex *pindexStart = pindexBestHeader->pprev ? pindexBestHeader->pprev : pindexBestHeader;
const CBlockIndex *pindexStart = pindexBestHeader;
/* If possible, start at the block preceding the currently
best known header. This ensures that we always get a
non-empty list of headers back as long as the peer
is up-to-date. With a non-empty response, we can initialise
the peer's known best block. This wouldn't be possible
if we requested starting at pindexBestHeader and
got back an empty response. */
if (pindexStart->pprev)
pindexStart = pindexStart->pprev;
LogPrint("net", "initial getheaders (%d) to peer=%d (startheight:%d)\n", pindexStart->nHeight, pto->id, pto->nStartingHeight);
pto->PushMessage("getheaders", chainActive.GetLocator(pindexStart), uint256());
}