This patches silences the remaining discarded future warnings, those
where it cannot be determined with reasonable confidence that this was
indeed the actual intent of the author, or that the discarding of the
future could lead to problems. For all those places a FIXME is added,
with the intent that these will be soon followed-up with an actual fix.
I deliberately haven't fixed any of these, even if the fix seems
trivial. It is too easy to overlook a bad fix mixed in with so many
mechanical changes.
Explicity pass the port number of the AWS metadata server API
when creating a corresponding socket.
This patch fixes the regression introduced by 4ef940169f.
Fixes#4719
Signed-off-by: Vlad Zolotarov <vladz@scylladb.com>
Rather than {std::experimental,boost,seastar::compat}::filesystem
On Sat, 2019-03-23 at 01:44 +0200, Avi Kivity wrote:
> The intent for seastar::compat was to allow the application to choose
> the C++ dialect and have seastar follow, rather than have seastar choose
> the types and have the application follow (as in your patch).
Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
Replace stdx::optional and stdx::string_view with the C++ std
counterparts.
Some instances of boost::variant were also replaced with std::variant,
namely those that called seastar::visit.
Scylla now requires GCC 8 to compile.
Signed-off-by: Duarte Nunes <duarte@scylladb.com>
Message-Id: <20190108111141.5369-1-duarte@scylladb.com>
"
=== How the the partition level repair works
- The repair master decides which ranges to work on.
- The repair master splits the ranges to sub ranges which contains around 100
partitions.
- The repair master computes the checksum of the 100 partitions and asks the
related peers to compute the checksum of the 100 partitions.
- If the checksum matches, the data in this sub range is synced.
- If the checksum mismatches, repair master fetches the data from all the peers
and sends back the merged data to peers.
=== Major problems with partition level repair
- A mismatch of a single row in any of the 100 partitions causes 100
partitions to be transferred. A single partition can be very large. Not to
mention the size of 100 partitions.
- Checksum (find the mismatch) and streaming (fix the mismatch) will read the
same data twice
=== Row level repair
Row level checksum and synchronization: detect row level mismatch and transfer
only the mismatch
=== How the row level repair works
- To solve the problem of reading data twice
Read the data only once for both checksum and synchronization between nodes.
We work on a small range which contains only a few mega bytes of rows,
We read all the rows within the small range into memory. Find the
mismatch and send the mismatch rows between peers.
We need to find a sync boundary among the nodes which contains only N bytes of
rows.
- To solve the problem of sending unnecessary data.
We need to find the mismatched rows between nodes and only send the delta.
The problem is called set reconciliation problem which is a common problem in
distributed systems.
For example:
Node1 has set1 = {row1, row2, row3}
Node2 has set2 = { row2, row3}
Node3 has set3 = {row1, row2, row4}
To repair:
Node1 fetches nothing from Node2 (set2 - set1), fetches row4 (set3 - set1) from Node3.
Node1 sends row1 and row4 (set1 + set2 + set3 - set2) to Node2
Node1 sends row3 (set1 + set2 + set3 - set3) to Node3.
=== How to implement repair with set reconciliation
- Step A: Negotiate sync boundary
class repair_sync_boundary {
dht::decorated_key pk;
position_in_partition position
}
Reads rows from disk into row buffers until the size is larger than N
bytes. Return the repair_sync_boundary of the last mutation_fragment we
read from disk. The smallest repair_sync_boundary of all nodes is
set as the current_sync_boundary.
- Step B: Get missing rows from peer nodes so that repair master contains all the rows
Request combined hashes from all nodes between last_sync_boundary and
current_sync_boundary. If the combined hashes from all nodes are identical,
data is synced, goto Step A. If not, request the full hashes from peers.
At this point, the repair master knows exactly what rows are missing. Request the
missing rows from peer nodes.
Now, local node contains all the rows.
- Step C: Send missing rows to the peer nodes
Since local node also knows what peer nodes own, it sends the missing rows to
the peer nodes.
=== How the RPC API looks like
- repair_range_start()
Step A:
- request_sync_boundary()
Step B:
- request_combined_row_hashes()
- reqeust_full_row_hashes()
- request_row_diff()
Step C:
- send_row_diff()
- repair_range_stop()
=== Performance evaluation
We created a cluster of 3 Scylla nodes on AWS using i3.xlarge instance. We
created a keyspace with a replication factor of 3 and inserted 1 billion
rows to each of the 3 nodes. Each node has 241 GiB of data.
We tested 3 cases below.
1) 0% synced: one of the node has zero data. The other two nodes have 1 billion identical rows.
Time to repair:
old = 87 min
new = 70 min (rebuild took 50 minutes)
improvement = 19.54%
2) 100% synced: all of the 3 nodes have 1 billion identical rows.
Time to repair:
old = 43 min
new = 24 min
improvement = 44.18%
3) 99.9% synced: each node has 1 billion identical rows and 1 billion * 0.1% distinct rows.
Time to repair:
old: 211 min
new: 44 min
improvement: 79.15%
Bytes sent on wire for repair:
old: tx= 162 GiB, rx = 90 GiB
new: tx= 1.15 GiB, tx = 0.57 GiB
improvement: tx = 99.29%, rx = 99.36%
It is worth noting that row level repair sends and receives exactly the
number of rows needed in theory.
In this test case, repair master needs to receives 2 million rows and
sends 4 million rows. Here are the details: Each node has 1 billion *
0.1% distinct rows, that is 1 million rows. So repair master receives 1
million rows from repair slave 1 and 1 million rows from repair slave 2.
Repair master sends 1 million rows from repair master and 1 million rows
received from repair slave 1 to repair slave 2. Repair master sends
sends 1 million rows from repair master and 1 million rows received from
repair slave 2 to repair slave 1.
In the result, we saw the rows on wire were as expected.
tx_row_nr = 1000505 + 999619 + 1001257 + 998619 (4 shards, the numbers are for each shard) = 4'000'000
rx_row_nr = 500233 + 500235 + 499559 + 499973 (4 shards, the numbers are for each shard) = 2'000'000
Fixes: #3033
Tests: dtests/repair_additional_test.py
"
* 'asias/row_level_repair_v7' of github.com:cloudius-systems/seastar-dev: (51 commits)
repair: Enable row level repair
repair: Add row_level_repair
repair: Add docs for row level repair
repair: Add repair_init_messaging_service_handler
repair: Add repair_meta
repair: Add repair_writer
repair: Add repair_reader
repair: Add repair_row
repair: Add fragment_hasher
repair: Add decorated_key_with_hash
repair: Add get_random_seed
repair: Add get_common_diff_detect_algorithm
repair: Add shard_config
repair: Add suportted_diff_detect_algorithms
repair: Add repair_stats to repair_info
repair: Introduce repair_stats
flat_mutation_reader: Add make_generating_reader
storage_service: Introduce ROW_LEVEL_REPAIR feature
messaging_service: Add RPC verbs for row level repair
repair: Export the repair logger
...
Fixes#2896 (hopefully)
Implementation of origin change c000da13563907b99fe220a7c8bde3c1dec74ad5
Reduces the amount of maps and sets and general complexity of
endpoint calculation by simply mapping dc:s to expected node
counts, re-using endpoint sets and iterate thusly.
Tested with transposed origin unit test comparing old vs. new
algo results. (Next patch)
* seastar d59fcef...b924495 (2):
> build: Fix protobuf generation rules
> Merge "Restructure files" from Jesse
Includes fixup patch from Jesse:
"
Update Seastar `#include`s to reflect restructure
All Seastar header files are now prefixed with "seastar" and the
configure script reflects the new locations of files.
Signed-off-by: Jesse Haber-Kucharsky <jhaberku@scylladb.com>
Message-Id: <5d22d964a7735696fb6bb7606ed88f35dde31413.1542731639.git.jhaberku@scylladb.com>
"
Public IP is required for Ec2MultiRegionSnitch. If it's not available
different snitch should be used.
This patch would result in a readable error message to be printed
instead of just a cryptic message with HTTP response body.
Fixes#3897
Signed-off-by: Vlad Zolotarov <vladz@scylladb.com>
sprint() recently became more strict, throwing on sprint("%s", 5). Replace
with the more modern format().
Mechanically converted with https://github.com/avikivity/unsprint.
get_ranges() is supposed to return ranges in sorted order. However, a35136533d
broke this and returned the range that was supposed to be last in the second
position (e.g. [0, 10, 1, 2, 3, 4, 5, 6, 7, 8, 9]). The broke cleanup, which
relied on the sort order to perform a binary search. Other users of the
get_ranges() family did not rely on the sort order.
Fixes#3872.
Message-Id: <20181019113613.1895-1-avi@scylladb.com>
In commit 4a0b561376, "storage_service:
Get rid of moving operation", we removed remove_from_moving() in
update_normal_tokens(). However, remove_from_moving() calls
invalidate_cached_rings(). We should call invalidate_cached_rings() in
update_normal_tokens(), otherwise we will get wrong token range to
address map in the token_metadata cache.
This issue exists in master only. It is not in any of the releases.
Message-Id: <c03f2ed478cfdb84494f36dce9a8cfc05ed9e0cd.1538288364.git.asias@scylladb.com>
We need the mapping between dht::token_range to
std::vector<inet_address> and inet_address to dht::token_range_vector in
various places. Currently, we use std::unordered_multimap and convert to
std::unordered_map. It is better to use std::unordered_map in the first
place. The changes like below:
- Change from
std::unordered_multimap<dht::token_range, inet_address>
to
std::unordered_map<dht::token_range, std::vector<inet_address>>
- Change from
std::unordered_multimap<inet_address, dht::token_range>
to
std::unordered_map<inet_address, dht::token_range_vector>
Message-Id: <b8ecc41775e46ec064db3ee07510c404583390aa.1533106019.git.asias@scylladb.com>
The moving operation changes a node's token to a new token. It is
supported only when a node has one token. The legacy moving operation is
useful in the early days before the vnode is introduced where a node has
only one token. I don't think it is useful anymore.
In the future, we might support adjusting the number of vnodes to reblance
the token range each node owns.
Removing it simplifies the cluster operation logic and code.
Fixes#3475
Message-Id: <144d3bea4140eda550770b866ec30e961933401d.1533111227.git.asias@scylladb.com>
When nodetool repair is used with the combination of the "-pr" (primary
range) and "-local" (only repair with nodes in the same DC) options,
Scylla needs to define the "primary ranges" differently: Rather than
assign one node in the entire cluster to be the primary owner of every
token, we need one node in each data-center - so that a "-local"
repair will cover all the tokens.
Fixes#3557.
Signed-off-by: Nadav Har'El <nyh@scylladb.com>
Message-Id: <20180701132445.21685-1-nyh@scylladb.com>
ec2_snitch::gossiper_starting() calls for the base class (default) method
that sets _gossip_started to TRUE and thereby prevents to following
reconnectable_snitch_helper registration.
Fixes#3454
Signed-off-by: Vlad Zolotarov <vladz@scylladb.com>
Message-Id: <1528208520-28046-1-git-send-email-vladz@scylladb.com>
This patch implements the last_replicas returning part of the query()
signature changes for singular queries. It allows for client code to
save the last returned replicas and pass it to query() on the next page
as the preferred-replicas parameter, thus faciliate the read requests
for the next page hitting the same replicas.
De-inlining allows us to remove some dependencies, and those functions
are too complex to inline anyway.
A few always-throwing functions get the [[noreturn]] attribute to
avoid damaging code generation.
"This series adds the GoogleCloudSnitch.
Fixes#1619"
* 'google-cloud-snitch-v4' of https://github.com/vladzcloudius/scylla:
config: uncomment/add the supported snitches description
tests: added gce_snitch_test
locator::gce_snitch: implementation of the GoogleCloudSnitch
locator::snitch_base: properly log the failure during the snitch startup
affected_ranges can be very large in a large cluster or node with big
num_tokens account. calculate_natural_endpoints takes more time to
process in this case as well.
Futurize calculate_pending_ranges_for_leaving and handle the loop with
do_for_each to give some time for the reactor to breath, so it does not
block.
token_metadata::calculate_pending_ranges is a complicated function.
Split it into 3 parts for leaving operation, moving opeartion,
bootstrap opeartion.
Now, do_update_pending_ranges is futurized. We can finally futurize
token_metadata::calculate_pending_ranges in order to convert the loops
inside it to do_for_each insead of plain for loops to avoid reactor
stall.
This is a snitch that should be used when Scylla runs in GCE VMs in both
single and multi data center (DC) configurations.
This snitch interacts with the GCE (instance metadata) API as
described here: https://cloud.google.com/compute/docs/storing-retrieving-metadata)
similarly to how ec2_snitchXXX interacts with the AWS API.
However unlike ec2_multi_region_snitch the GCE snitch only gets the instance's zone and sets
the DC and the RACK based on it, e.g. for us-central1-a the DC is set to 'us-central'
and the RACK - to 'a'.
GCE snitch doesn't have to learn the internal and external IPs of the instance because in
GCE instances from different regions can interact using internal IPs (in the AWS they can't).
Signed-off-by: Vlad Zolotarov <vladz@scylladb.com>
gossiper::get_endpoint_state_for_endpoint() returns a copy of
endpoint_state, which we've seen can be very expensive.
This patch adds a similar function which returns a pointer instead,
and changes the call sites where using the pointer-returning variant
is deemed safe (the pointer neither escapes the function, nor crosses
any defer point).
Fixes#764
Signed-off-by: Duarte Nunes <duarte@scylladb.com>
Make get_natural_endpoints return local address iff token metadata
is not yet setup (since that is the one address we already know of).
If a request has a consistency level requiring more endpoints, it
will still fail, but for calls with, for example, CL=ONE, at startup
we will succeed, and more or less act like local strategy. Yet,
further down the line, have data distributed as desired.
Acked-by: Gleb Natapov <gleb@scylladb.com>
Message-Id: <20170926113512.15707-1-calle@scylladb.com>
- introcduced "seastarx.hh" header, which does a "using namespace seastar";
- 'net' namespace conflicts with seastar::net, renamed to 'netw'.
- 'transport' namespace conflicts with seastar::transport, renamed to
cql_transport.
- "logger" global variables now conflict with logger global type, renamed
to xlogger.
- other minor changes
Caught by PVS-Studio static analyzer:
The object was created but it is not being used. The 'throw' keyword could be missing: throw runtime_error(FOO);
Reported-by: Phillip Khandeliants
Signed-off-by: Vlad Zolotarov <vladz@scylladb.com>
Wrapping ranges are a pain, so we are moving wrap handling to the edges.
Since cql can't generate wrapping ranges, this means thrift and the ring
maintenance code; also range->ring transformations need to merge the first
and last ranges.
Message-Id: <1478105905-31613-1-git-send-email-avi@scylladb.com>
