Sometimes SNOs fail to properly configure or lose connection to their storage directory
which can result in DQ. This causes unnecessary repair and is unfortunate for all parties.
This change introduces the creation of a special file in the storage directory at runtime
containing the node ID. While the storage node runs, it periodically verifies that it can
find said file with the correct contents in the correct location. If not, the node will
shut down with an error message.
This change will solve the issue of nodes losing access to the storage directory, but it will not
solve the issue of nodes pointing to the wrong directory, as the identifying file is created each
time the node starts up. After this change has been the minimum version for a few releases, we will
remove the creation of the directory-identifying file from the storage node run command and add it
to the setup command.
Change-Id: Ib7b10e96ac07373219835e39239e93957e7667a4
To prevent storagenode from implicitly recreating missing dbs and storage,
as such behaviour leads to audit failures. Do not allow storagenode to
start if any of dbs or storage is missing, corrupted, or dedicated storage disk is
unmounted, to get downtime instead.
Change-Id: Ic64e1f0ff4d8ef5b2fddbe7a7e53df4f4bd8652e
What:
Use the github.com/jackc/pgx postgresql driver in place of
github.com/lib/pq.
Why:
github.com/lib/pq has some problems with error handling and context
cancellations (i.e. it might even issue queries or DML statements more
than once! see https://github.com/lib/pq/issues/939). The
github.com/jackx/pgx library appears not to have these problems, and
also appears to be better engineered and implemented (in particular, it
doesn't use "exceptions by panic"). It should also give us some
performance improvements in some cases, and even more so if we can use
it directly instead of going through the database/sql layer.
Change-Id: Ia696d220f340a097dee9550a312d37de14ed2044
This runs each benchmark for one iteration to ensure that they are
valid. Unfortunately, it does not give any useful metrics as output.
Change-Id: I68940398c8dd849aed656bd12656f48d5df10128
errors.New errors will not show up in monkit tracing
as a useful error type. this change fixes a test (!)
and makes it so monkit will tell us what the error
type is, if we have this failure
Change-Id: Ic9933704e4095495c7ee286d9df3eb7eb94b25c9
A file piece could be deleted in between walking the list of files read
from a directory and before we actually perform any operation on such
file. When that happens, we don't want to return an error, we want to
just ignore it and carry on.
Change-Id: I8f6986070e5883599a08fccf8b125c075b30fe1b
This ensures that rows are closed to avoid leaks.
Also verifies that Err() is called, to ensure that no
error is left behind.
Change-Id: Idd1bec9bf479f40021da67b2c80ce83033149469
It looks like GetAll and DeleteMultiple are only used in tests for now,
but they didn't have handling for retry errors returned from cockroach.
If they're used in prod in the future, now they will retry.
Change-Id: I0f281454ddebf282789142ff1d66a69bda5727c9
In walkNamespaceWithPrefix log in case of "lstat" error, because this may indicate an underlying disk corruption.
SG-50
Change-Id: I867c3ffc47cfac325ae90658ec4780d213ff3e63
This changeset replaces https://review.dev.storj.io/c/storj/storj/+/1839
which did the same thing but Nat couldn't figure out how to fix conflicting
files the correct gerrity way.
Change-Id: If05a8902aca986ea9f6c9168a90b31beebab839a
Currently uploads can cause a lot of IOPS, reduce this by introducing a
in-memory buffer on-top of the file.
Change-Id: I5f4e3e01c0a36258271d180b922107de447bcb59
Currently Cockroach isn't performant for concurrent database setup and
tear-down. Instead of a single instance allow setting multiple potential
connection strings and let the tests pick one connection string
randomly.
This improves test duration by ~10 minutes.
While we are at significantly changing how pgtest works, introduce
helper PickPostgres and PickCockroach for selecting the database to
reduce code duplications in multiple places.
Change-Id: I8ad171d5c4c8a4fc081ec2ae9bdd0cc948a80619
In cases like the segment reaper script connecting to the metainfodb,
we don't want a db migration to happen automatically when we call
metainfo.NewStore. This adds MigrateToLatest method for postgreskv
and cockroackv, and calls MigrateToLatest in places where NewStore used
to create tables.
Change-Id: I682d0f26d609af0601dfdb32a24866cdf5d32a7e
Currently ListV2 loaded the whole data into memory, even when all the
data wasn't being used, using up more memory than needed.
Change-Id: I5846d979344729b447c108a6cc9f4227229ec981
storagenodes have like 10 or more databases. without this
tag they all get sent as the same value, stomping on each
other.
Change-Id: Ib12019684d6ea8f2a5b83df584056dfa79e3c4b3
this commit updates our monkit dependency to the v3 version where
it outputs in an influx style. this makes discovery much easier
as many tools are built to look at it this way.
graphite and rothko will suffer some due to no longer being a tree
based on dots. hopefully time will exist to update rothko to
index based on the new metric format.
it adds an influx output for the statreceiver so that we can
write to influxdb v1 or v2 directly.
Change-Id: Iae9f9494a6d29cfbd1f932a5e71a891b490415ff
DeleteMultiple will allow metainfo to delete multiple segments
and get the old pointers in a single request.
Change-Id: Ic144f30c5453274fa2b80df2895f123f5a9cc48b
Currently storage tests were tied to the default lookup limit.
By increasing the limits, the tests will take longer and sometimes
cause a large number of goroutines to be started.
This change adds configurable lookup limit to all storage backends.
Also remove boltdb.NewShared, since it's not used any more.
Change-Id: I1a052f149da471246fac5745da133c3cfc27582e
COUNT on a SCAN does not actually limit the results [1]. It limits the
amount of work a single call to SCAN will perform before returning. By
setting this to limit we can sometimes timeout on the request if limit
is very large.
This restores storage/redis back to it's original behavior.
[1]: https://redis.io/commands/scan#the-count-option
Change-Id: Ia75afb5152df909df38c9a7c6feb74d062f49d6a
This was inadvertently converted to the Cockroach version. This reverts
most of that and keeps the changes since then.
Change-Id: Ia440eeebb01bc89fbfa8ce266668030173061469
Replace all the remaining uses of sql.DB with tagsql.DB to
fix issues with context cancellation.
Introduce tagsql.Open which helps to get rid of all tagsql.Wrap-s.
Use tagsql in cockroachkv and postgreskv.
Change-Id: I8946d203341cb85a25976896fc7881e1f704e779
* Plumbs the limit through all backends ensuring they don't do
unnecessary work.
* Don't arbitrarily limit at the backend with hardcoded defaults. The
limit will be set by the caller.
Prior to this change the code on recursive in some backends would do 10k
results from the database and then only return the first 1k (throwing
out 9k of them).
Prior to this change some backends had no limit at all (e.g. redis).
Change-Id: I1f327eefe095776d123dd11362cd00994c22efdf
