Commit Graph

49 Commits

Author SHA1 Message Date
Wenbiao Zheng 6810933640 eth/downloader: SetBlocksIdle is not used (#17962)
__
 <(o )___
  ( ._> /
   `---'
2018-10-24 01:27:49 +02:00
Yusup 7aad81f881 eth: fix typos (#16414) 2018-04-04 12:25:02 +02:00
Ricardo Domingos 9ff9d04a69 all: fix code comment typos (#15547)
* console: fix typo in comment

* contracts/release: fix typo in comment

* core: fix typo in comment

* eth: fix typo in comment

* miner: fix typo in comment
2017-11-24 11:20:01 +02:00
Péter Szilágyi f46adfac28
eth/downloader: track peer drops and deassign state sync tasks 2017-09-12 15:13:14 +03:00
Nick Johnson fe13949d9d eth/downloader: Doc fixes 2017-07-05 11:13:16 +01:00
Nick Johnson be8f8409bc eth/downloader, les, light: Changes in response to review 2017-07-03 15:17:12 +01:00
Nick Johnson ae11545bc5 eth, les: Refactor downloader peer to use structs 2017-06-29 12:49:18 +01:00
Felix Lange 0042f13d47 eth/downloader: separate state sync from queue (#14460)
* eth/downloader: separate state sync from queue

Scheduling of state node downloads hogged the downloader queue lock when
new requests were scheduled. This caused timeouts for other requests.
With this change, state sync is fully independent of all other downloads
and doesn't involve the queue at all.

State sync is started and checked on in processContent. This is slightly
awkward because processContent doesn't have a select loop. Instead, the
queue is closed by an auxiliary goroutine when state sync fails. We
tried several alternatives to this but settled on the current approach
because it's the least amount of change overall.

Handling of the pivot block has changed slightly: the queue previously
prevented import of pivot block receipts before the state of the pivot
block was available. In this commit, the receipt will be imported before
the state. This causes an annoyance where the pivot block is committed
as fast block head even when state downloads fail. Stay tuned for more
updates in this area ;)

* eth/downloader: remove cancelTimeout channel

* eth/downloader: retry state requests on timeout

* eth/downloader: improve comment

* eth/downloader: mark peers idle when state sync is done

* eth/downloader: move pivot block splitting to processContent

This change also ensures that pivot block receipts aren't imported
before the pivot block itself.

* eth/downloader: limit state node retries

* eth/downloader: improve state node error handling and retry check

* eth/downloader: remove maxStateNodeRetries

It fails the sync too much.

* eth/downloader: remove last use of cancelCh in statesync.go

Fixes TestDeliverHeadersHang*Fast and (hopefully)
the weird cancellation behaviour at the end of fast sync.

* eth/downloader: fix leak in runStateSync

* eth/downloader: don't run processFullSyncContent in LightSync mode

* eth/downloader: improve comments

* eth/downloader: fix vet, megacheck

* eth/downloader: remove unrequested tasks anyway

* eth/downloader, trie: various polishes around duplicate items

This commit explicitly tracks duplicate and unexpected state
delieveries done against a trie Sync structure, also adding
there to import info logs.

The commit moves the db batch used to commit trie changes one
level deeper so its flushed after every node insertion. This
is needed to avoid a lot of duplicate retrievals caused by
inconsistencies between Sync internals and database. A better
approach is to track not-yet-written states in trie.Sync and
flush on commit, but I'm focuing on correctness first now.

The commit fixes a regression around pivot block fail count.
The counter previously was reset to 1 if and only if a sync
cycle progressed (inserted at least 1 entry to the database).
The current code reset it already if a node was delivered,
which is not stong enough, because unless it ends up written
to disk, an attacker can just loop and attack ad infinitum.

The commit also fixes a regression around state deliveries
and timeouts. The old downloader tracked if a delivery is
stale (none of the deliveries were requestedt), in which
case it didn't mark the node idle and did not send further
requests, since it signals a past timeout. The current code
did mark it idle even on stale deliveries, which eventually
caused two requests to be in flight at the same time, making
the deliveries always stale and mass duplicating retrievals
between multiple peers.

* eth/downloader: fix state request leak

This commit fixes the hang seen sometimes while doing the state
sync. The cause of the hang was a rare combination of events:
request state data from peer, peer drops and reconnects almost
immediately. This caused a new download task to be assigned to
the peer, overwriting the old one still waiting for a timeout,
which in turned leaked the requests out, never to be retried.
The fix is to ensure that a task assignment moves any pending
one back into the retry queue.

The commit also fixes a regression with peer dropping due to
stalls. The current code considered a peer stalling if they
timed out delivering 1 item. However, the downloader never
requests only one, the minimum is 2 (attempt to fine tune
estimated latency/bandwidth). The fix is simply to drop if
a timeout is detected at 2 items.

Apart from the above bugfixes, the commit contains some code
polishes I made while debugging the hang.

* core, eth, trie: support batched trie sync db writes

* trie: rename SyncMemCache to syncMemBatch
2017-06-22 15:26:03 +03:00
Péter Szilágyi 9184249b39 Logger updates 3 (#3730)
* accounts, cmd, eth, ethdb: port logs over to new system

* ethdb: drop concept of cache distribution between dbs

* eth: fix some log nitpicks to make them nicer
2017-03-02 14:06:16 +01:00
Péter Szilágyi 0a63c3e362
eth/downloader: port over old logs from glog to log15 2017-02-27 13:16:40 +02:00
Felix Lange 35a7dcb162 all: gofmt -w -s 2017-01-06 15:52:03 +01:00
Péter Szilágyi 1dd272080d eth, eth/downloader: better remote head tracking 2016-07-25 15:14:14 +03:00
Felix Lange 016007bd25 eth, eth/downloader, eth/fetcher: delete eth/61 code
The eth/61 protocol was disabled in #2776, this commit removes its
message handlers and hash-chain sync logic.
2016-07-22 13:17:19 +02:00
Péter Szilágyi 88f174a014 eth/downloader: adaptive quality of service tuning 2016-06-06 14:20:57 +03:00
Péter Szilágyi b40dc8a1da eth/downloader: implement concurrent header downloads 2016-05-17 10:03:34 +03:00
Péter Szilágyi e3f2b541f2 eth/downloader: fix a throughput estimation data race 2016-03-10 12:58:46 +02:00
Péter Szilágyi b6f5523bdc eth/downloader: fetch data proportionally to peer capacity 2015-11-19 17:01:39 +02:00
Péter Szilágyi b658a73ed5 eth/downloader: fix dysfunctional ignore list hidden by generic set 2015-11-04 13:11:52 +02:00
Péter Szilágyi 5b0ee8ec30 core, eth, trie: fix data races and merge/review issues 2015-10-21 16:49:55 +03:00
Péter Szilágyi ab27bee25a core, eth, trie: direct state trie synchronization 2015-10-19 10:03:09 +03:00
Péter Szilágyi f186b39018 eth/downloader: add fast and light sync strategies 2015-10-19 10:03:09 +03:00
Péter Szilágyi 47f62a67aa eth/downloader: match capabilities when querying idle peers 2015-10-02 13:20:41 +03:00
Péter Szilágyi 47a7fe5d22 eth: port the synchronisation algo to eth/62 2015-08-25 17:48:47 +03:00
Felix Lange bfbcfbe4a9 all: fix license headers one more time
I forgot to update one instance of "go-ethereum" in commit 3f047be5a.
2015-07-23 18:35:11 +02:00
Felix Lange 3f047be5aa all: update license headers to distiguish GPL/LGPL
All code outside of cmd/ is licensed as LGPL. The headers
now reflect this by calling the whole work "the go-ethereum library".
2015-07-22 18:51:45 +02:00
Felix Lange ea54283b30 all: update license information 2015-07-07 14:12:44 +02:00
Péter Szilágyi f43c07cb3c eth, eth/downloader: transition to eth 61 2015-06-30 19:05:06 +03:00
Péter Szilágyi af51dc4d63 eth, eth/downloader: pass the eth protocol version through 2015-06-30 19:00:01 +03:00
Péter Szilágyi b240983e2b eth, eth/downloader: do async block fetches, add dl tests 2015-06-15 12:26:05 +03:00
Péter Szilágyi 63c6cedb14 eth/downloader: cap the hash ban set, add test for it 2015-06-08 14:12:00 +03:00
Péter Szilágyi 94e4aa6ea9 eth/downloader: log hard timeouts and reset capacity 2015-06-05 11:53:46 +03:00
Péter Szilágyi d754c25cc8 eth/downloader: drop log entry from peer, it's covered already 2015-06-04 16:22:55 +03:00
Péter Szilágyi 28c32d1b1b eth/downloader: fix #1178, don't request blocks beyond the cache bounds 2015-06-04 14:51:14 +03:00
Péter Szilágyi 3ec159ab6b eth/downloader: demote peers if they exceed the soft limits at 1 blocks already 2015-06-03 15:43:12 +03:00
Péter Szilágyi c9a546c310 eth/downloader: add a basic block download congestion control 2015-06-03 14:40:11 +03:00
Péter Szilágyi fa53c5e074 eth/downloader: use count instead of peers, clearer 2015-05-11 17:10:00 +03:00
Péter Szilágyi ebbd8b0743 eth/downloader: revert to demotion, use harsher penalty 2015-05-11 16:47:58 +03:00
Péter Szilágyi 685862d2ce eth/downloader: fix #910, thread safe peers & polishes 2015-05-11 14:26:20 +03:00
Péter Szilágyi 4800c94392 eth/downloader: prioritize block fetch based on chain position, cap memory use 2015-05-07 10:57:34 +03:00
obscuren bd9c76097d eth/downloader: removed peer td management and best peer selection 2015-04-24 15:37:07 +02:00
obscuren c58918c84a downloader: moved chunk ignoring. Fixes issue with catching up 2015-04-19 13:30:34 +02:00
obscuren 5c59d95532 downloader: defer peer reset after download 2015-04-19 01:39:37 +02:00
obscuren a1d97ea4db typo 2015-04-18 20:35:49 +02:00
obscuren 6830ddb659 downloader: free up peers from work when the downloader resets 2015-04-18 20:25:55 +02:00
obscuren c2c24b3bb4 downloader: improved downloading and synchronisation
* Downloader's peers keeps track of peer's previously requested hashes
  so that we don't have to re-request
* Changed `AddBlock` to be fully synchronous
2015-04-18 18:55:13 +02:00
obscuren 205378016f downloader: added demotion / promotion in prep. for rep. system 2015-04-16 02:16:33 +02:00
obscuren a8a2b2a488 downloader: added missing blocks catchup functionality
When a parent is missing in the block list an attempt should be made to
fetch the missing parent and grandparents.
2015-04-13 16:38:32 +02:00
obscuren 7dcb9825c3 downloader: return an error for peer.fetch and return chunk to queue
If a peer was somehow already fetching and somehow managed to end up in
the `available` pool it should return it's work.
2015-04-12 13:36:01 +02:00
obscuren acf8452c33 downloader: implemented new downloader 2015-04-12 12:38:25 +02:00