The previous sha was taken from a revision on a devel branch that I
thought would continue staying in the tree after it was merged. That
revision was rebased away and the API was changed for the logger.
This updates the usage of the logger and adds a simple package for
constructing the base logger.
The 1.0 version of zap changed the format of the default console logger
so this change moves over to this new logger instead of attempting to
retain backwards compatibility with the old format.
There was a very small window where it was possible to deadlock during
the close of the Store. When closing, the Store waited on its Waitgroup
under a `Lock`. Naturally, all other goroutines must have been in a
position to call `Done` on the `Waitgroup` before the `Wait` call in
`Close` would return.
For the goroutine running the `monitorShards` method it was possible
that it would be unable to do this. Specifically, if the `monitorShards`
goroutine was jumping into the `t.C` case as the `Close()` goroutine was
acquiring the `Lock` then then `monitorShards` goroutine would be unable
to acquire the `RLock`. Since it would also be unable to progress around
its loop to jump into the `s.closing` case, it would be unable to call
`Done` on the `WaitGroup` and we would have a deadlock.
This was identified during an AppVeyor CI run, though I was unable to
reproduce this locally.
Previously we used the EngineOptions to determine which shard index
type we were using. However, these options are set once at runtime
initialisation. Therefore if you're running with TSI enabled but then
accessing a legacy database with the inmem index, TagValues would not
have taken advantage of the inmem index.
This change ensures we always check the actual index of the shard(s).
This commit adds time support to SHOW TAG VALUES. Time can be used as
both a lower and upper boundary. However, there are some caveats.
For the `inmem` index, filtering by time will still return all results
because the index data is shared across shards.
For the `tsi1` index, filtering by time will only work down to the shard
lever. Specifically, when querying by time all shards within that time
range will be used to generate the results.
This changes the compaction scheduling to better utilize the available
cores that are free. Previously, a level was planned in its own goroutine
and would kick off a number of compactions groups. The problem with this
model was that if there were 4 groups, and 3 completed quickly, the planning
would be blocked for that level until the last group finished. If the compactions
at the prior level are running more quickly, a large backlog could accumlate.
This now moves the planning to a single goroutine that plans each level in
succession and starts as many groups as it can. When one group finishes,
the planning will start the next group for the level.
This instructs the kernel that it can release memory used by mmap'd
TSM files when they are not actively being used. It the mappings are
use, the kernel will fault the pages back in. On linux, this causes
RES memory to drop immediately when run.
A cold shard that suddenly receives a lot of writes could get a very
big cache that takes a long time to snapshot or causes the cache
max memory limit to be hit more quickly. This re-enables the compactions
if necessary during writes so we don't have to wait for the shard monitor
goroutine to re-enable them.
The ConditionExpr function is more accurate because it parses the
condition and ensures that time conditions are actually used correctly.
That means that attempting to combine conditions with OR will not result
in the query silently pretending it's an AND and nested conditions work
correctly so there is only one way to read the query.
It also extracts the non-time conditions into a separate condition so we
can stop attempting to parse around the time conditions in lower layers
of the storage engine. This change does not remove those hacks, but a
following commit should be able to sanitize the condition and remove
them.
The tag cardinality checks were run for all inmem shards. Since inmem
shards share the same index, a lot of the work is redundant. Inmem shards
also need to sort their measurmenet and tag keys which can be CPU intensive
with many shards or higher cardinality.
This changes the monitoring to just check one shard in each database which
should lower CPU usage due to excessive sorting. The longer term solution
is to use TSI which would not have this check or required sorting.
There was a change to speed up deleting and dropping measurements
that executed the deletes in parallel for all shards at once. #7015
When TSI was merged in #7618, the series keys passed into Shard.DeleteMeasurement
were removed and were expanded lower down. This causes memory to blow up
when a delete across many shards occurs as we now expand the set of series
keys N times instead of just once as before.
While running the deletes in parallel would be ideal, there have been a number
of optimizations in the delete path that make running deletes serially pretty
good. This change just limits the concurrency of the deletes which keeps memory
more stable.
When monitoring shards, a slice of measurements is allocated for
each shard. With many shards and measurements, these allocations
can be large. Since inmem shards share the same index, we only
need to do this once since the resulting slices are all the same.
This reduces memory usage when monitoring shard cardinality.
The monitor goroutine ran for each shard and updated disk stats
as well as logged cardinality warnings. This goroutine has been
removed by making the disks stats more lightweight and callable
direclty from Statisics and move the logging to the tsdb.Store. The
latter allows one goroutine to handle all shards.
Each shard has a number of goroutines for compacting different levels
of TSM files. When a shard goes cold and is fully compacted, these
goroutines are still running.
This change will stop background shard goroutines when the shard goes
cold and start them back up if new writes arrive.
This limit allows the number of concurrent level and full compactions
to be throttled. Snapshot compactions are not affected by this limit
as then need to run continously.
This limit can be used to control how much CPU is consumed by compactions.
The default is to limit to the number of CPU available.
Compactions are enabled as soon as the shard is opened. This can
slow down startup or cause the system to spike in CPU usage at startup
if many shards need to be compacted.
This now delays compactions until after they are loaded.
Removing series while trying to maintain the sorted series list
does not perform well when removing many series. This causes
drop DB, RP, series, to be very slow in some cases.
Instead, lazily create a sorted series list when first requested and
invalidate it when dropping series.
This reworks drop measurement to use a sorted list of series keys
instead of creating an intermediate map. It remove allocations
and some extra garbage that is created during drop measurement.
When using the inmem index, if one drops a database, and then creates it
again, the previous index object will be reused. This includes the
previous cardinality estimation sketches, leading to inaccurate
cardinality estimations.
They rebased a revision we were previously relying upon that allowed us
to use the vanity name so we are reverting back to an older version with
the old import path.
Adds a `tsdb.Series.ForEachTag()` function for safely iterating
over a series' tags within the context of a lock. This preverts
tags from being dereferenced during iteration which can cause
a seg fault.
This change adds some very basic name validation with the following
plain-english description: names must be non-zero sequence of printable
characters that do not contain slashes ('/' or '\') and are not equal to
either "." or "..".
The intent is that, since we currently just use database and retention
policy names directly as path elements, these rules will hopefully leave
us with names that should be at least close to valid directory names.
Ideally, we would restrict names even further or not use them as path
elements directly, but this should be a step towards the former without
restricting names "too much"
Fixes#7822
This change first ensures that databases and retention policies exist
before attempting to remove them from the Store. It also adds some
checks in the `DeleteDatabase` and `DeleteRetentionPolicy` to ensure
that maliciously named entries won't remove anything outside of the
configured data directory.
This adds query syntax support for subqueries and adds support to the
query engine to execute queries on subqueries.
Subqueries act as a source for another query. It is the equivalent of
writing the results of a query to a temporary database, executing
a query on that temporary database, and then deleting the database
(except this is all performed in-memory).
The syntax is like this:
SELECT sum(derivative) FROM (SELECT derivative(mean(value)) FROM cpu GROUP BY *)
This will execute derivative and then sum the result of those derivatives.
Another example:
SELECT max(min) FROM (SELECT min(value) FROM cpu GROUP BY host)
This would let you find the maximum minimum value of each host.
There is complete freedom to mix subqueries with auxiliary fields. The only
caveat is that the following two queries:
SELECT mean(value) FROM cpu
SELECT mean(value) FROM (SELECT value FROM cpu)
Have different performance characteristics. The first will calculate
`mean(value)` at the shard level and will be faster, especially when it comes to
clustered setups. The second will process the mean at the top level and will not
include that optimization.
It looks like the real import path to the project is go.uber.org/zap
instead of github.com/uber-go/zap since the example in the project
references that path.
The logging library has been switched to use uber-go/zap. While the
logging has been changed to use structured logging, this commit does not
change any of the logging statements to take advantage of the new
structured log or new log levels. Those changes will come in future
commits.
On my machine with about 20 shards, it would take 10+ seconds to shut
down InfluxDB with SIGINT. After this change, it shuts down in nearly
instantly.
(*tsdb.Store).Close was shutting down each of its shards sequentially.
Each shard's engine would signal to its compaction goroutines to quit,
and because each compaction goroutine has a hardcoded 1-second sleep in
between checks, waiting for the goroutines would often block for up to a
second.
This change closes all of the TSDB store's shards in parallel. This
means it's possible that multiple close values could error at once, but
we're still only returning the first error, consistent with previous
behavior. That being said, the return value of (*tsdb.Store).Close is
ignored in (*cmd/influxd/run.Server).Close anyway.
The FieldIterator is used to scan over the fields of a point, providing
information, and delaying parsing/decoding the value until it is needed.
This change uses this new type to avoid the allocation of a map for the
fields which is then thrown away as soon as the points get converted
into columns within the datastore.
When deleting a shard, the shard is locked and then removed from the
index. Removal from the index can be slow if there are a lot of
series. During this time, the shard is still expected to exist by
the meta store and tsdb store so stats collections, queries and writes
could all be run on this shard while it's locked. This can cause everything
to lock up until the unindexing completes and the shard can be unlocked.
Fixes#7226
This commit fixes the `MaxSelectSeriesN` limit which was broken by
the implementation of lazy iterators. The setting previously limited
the total number of series but the new implementation limits the
concurrent number of series being processed.
This commit limits queries to only process one shard at a time.
However, within a shard, multiple series can still be processed in
parallel. Shard iterators are lazily instantiated during query
execution to limit the amount of memory a given query uses.
The `SHOW MEASUREMENTS` and `SHOW TAG VALUES` cannot go through the
query engine to get the speed they need. They also only need access to
the database index and do not need access to specific shards. This
removes the query rewriting that was done to turn these two queries into
a select statement and reimplements them inside of the coordinator as an
interface on the TSDBStore.
A slower disk can can cause excessive allocations to occur when
writing to the WAL because the slower encoding and compression occurs
before taking the write lock. The encoding/compression grabs a large
byte slice from a pool and ultimately waits until it can acquire the
write lock.
This adds a throttle to limit how many inflight WAL writes can be queued
up to prevent OOMing the processess with slower disks and heavy writes.
Reduce the lock contention on tsdb.Store by taking a short lived
read-lock instead of a long write lock. Also close shards in parallel
and drop the whole RP dir in bulk instead of each shard dir.
Reduces the lock contention on the tsdb.Store by taking a short
read lock instead of a long write lock. Also processes shards
in parallel instead of serially.
Truncate the time interval output of the monitor service to be on even
time intervals rather than on every minute based on the start time. This
normalizes the output from the monitor service.
The tsdb package had a substantial amount of dead code related to the
old query engine still in there. It is no longer used, so it was removed
since it was left unmaintained. There is likely still more code that is
the same, but wasn't found as part of this code cleanup.
influxql has dead code show up because of the code generation so it is
not included in this pruning.
For restoring a shard, we need to be able to have the shard open,
but disabled. It was racy to open it and then disable it separately
since writes/queries could occur in between that time.
This locks showeed up in a deadlock systems running queries and
delete series across a large dataset. Queries should not need to
lock the tsdb.Store for writes
Drop database was closing and deleting each shard dir individually and
serially. It would then delete the empty database dirs.
This changes drop database to close all shards in parallel and run
one os.RemoveAll to remove everything under the db dir which is more
efficient.
This also reworked the locking to avoid locking the tsdb.Store for
long periods of time. That can cause queries and writes for other
databases to block as well.
This remove the dropMeta param from the tsdb.Store.DeleteSeries and
lets the shard determine when to remove the meta data from the index
based on what series still have data in the shard.
This uncovered a nasty bug in compactions where a fully deleted series would
prematurely end the compactions and not carry forward the rest of the data
in the TSM file. This is now fixed as well.
This has various benefits:
- Users embedding InfluxDB within other Go programs can specify a different logger / prefix easily.
- More consistent with code used elsewhere in InfluxDB (e.g. services, other `run.Server.*` fields, etc).
- This is also more efficient, because it means `executeQuery` no longer allocates a single `*log.Logger` each time it is called.
The QueryExecutor had a lot of dead code made obsolete by the query
engine refactor that has now been removed. The TSDBStore interface has
also been cleaned up so we can have multiple implementations of this
(such as a local and remote version).
A StatementExecutor interface has been created for adding custom
functionality to the QueryExecutor that may not be available in the open
source version. The QueryExecutor delegate all statement execution to
the StatementExecutor and the QueryExecutor will only keep track of
housekeeping. Implementing additional queries is as simple as wrapping
the cluster.StatementExecutor struct or replacing it with something
completely different.
The PointsWriter in the QueryExecutor has been changed to a simple
interface that implements the one method needed by the query executor.
This is to allow different PointsWriter implementations to be used by
the QueryExecutor. It has also been moved into the StatementExecutor
instead.
The TSDBStore interface has now been modified to contain the code for
creating an IteratorCreator. This is so the underlying TSDBStore can
implement different ways of accessing the underlying shards rather than
always having to access each shard individually (such as batch
requests).
Remove the show servers handling. This isn't a valid command in the open
source version of InfluxDB anymore.
The QueryManager interface is now built into QueryExecutor and is no
longer necessary. The StatementExecutor and QueryExecutor split allows
task management to much more easily be built into QueryExecutor rather
than as a separate struct.
Both Shard and Engine had the same reference to the measurementField map,
but they each protected it with their own locks. This causes a race when
write and queries are occurring because writes can add new fields to the
map while queries are reading from it.
The fix moves the ownership to the Engine and provides protected accessors
to that Shard now users. For the most parts, the access on shard were old
dead code.
Fixing the measurementFields map race created a new race on the internal
fields map. This is now unexported and protected via MeasurementFields
exported funcs.
Fixes#6188
Since loading a shard can allocate a lot of memory, running them all
at once could OOM the process. This limits the number of shards
loaded to 4. This will be changed to a config option provided the
approach helps.
This commit moves the `tsdb.Store.ExpandSources()` function onto
the `influxql.IteratorCreator` and provides support for issuing
source expansion across a cluster.
... by extracting the db/rp from the given path.
Now that the code has "standardized" on extracting db/rp this way, the
ShardLocation struct is no longer necessary and thus has been removed.
We're back on the previous style of passing the path and walPath to
NewShard.
This commit updates tsdb.Shard to contain a ShardConfig and updates
tsdb.Store to directly reference a map of tsdb.Shard rather than the
previous tsdb.shardLocation abstraction.
Fixes#5653 and #5394.
Previously dropping retention policies did not propogate to local TSDB
shards. Instead, the retention policiess would just be removed from the
Meta Store.
This PR adds ensures that data associated with retention policies is
removed, when the retention policy is dropped.
Also, it cleans up a couple of other methods in `tsdb`, including the
requirement to provide (redundant) shardIDs when deleting databases.
This commit removes `nil` shards returned from `tsdb.Store.Shards()`
which caused panics in some SELECTs. This can occur if the meta
store has created shards before the store or if the shards are
distributed throughout a cluster.
Fixes#5555
This changes backup and restore to work for TSM. It breaks it for b1 and bz1, but since those are getting removed it's ok.
The backup runs against any host that is specified and can backup either the metasstore, a database, specific retention policy, or a specific shard. It can also take incremental backups with the `since` flag, which will only backup TSM files that have been created since that timestamp.
The backup is safe to run online. However, for shards that are still hot for writes, they won't be able to create new TSM files while the backup for that single shard runs. If the backup isn't too large and the write throughput isn't too high this shouldn't be a problem since the writes will just go into the WAL cache.
Jonathan A. Sternberg
Ben Johnson
Edd Robinson
Stuart Carnie
Jason Wilder
Joe LeGasse
Mark Rushakoff
Mark Rushakoff
Cory LaNou
kun
Jason Wilder
Stephen Gutekanst
Justin Nuß
Paul Dix
David Norton