Right now, the best way for a storage node operator to get the current
space used for each satellite is to run the `storagenode exit-satellite`
command for graceful exit, and cancel at the second confirmation prompt.
This is convoluted and the data is readily available from the Blobs
Usage Cache.
This change adds the current space used by each satellite to the
endpoints `/api/sno` and `/api/sno/satellite/<Satellite ID>`
Change-Id: I2173005bb016fc76db96fd598d26b485e5b2aa0b
log.Fatal immediately terminates the program without running any defers.
We should properly close all the services and databases.
Change-Id: I5e959cef3eafedeacb3a2062e3da47e8d04e8e75
If we see an UnexpectedEOF error when attempting to read orders, return
the orders we have been able to read successfully and do not return an
error. This behavior ensures that the storagenode orders service
attempts to archive corrupted files and does not retry them repeatedly
and get stuck.
Change-Id: I0d00d1e174f968af6e99ca861eddad190f1339e2
This change accomplishes multiple things:
1. Instead of having a max in flight time, which means
we effectively have a minimum bandwidth for uploads
and downloads, we keep track of what windows have
active requests happening in them.
2. We don't double check when we save the order to see if it
is too old: by then, it's too late. A malicious uplink
could just submit orders outside of the grace window and
receive all the data, but the node would just not commit
it, so the uplink gets free traffic. Because the endpoints
also check for the order being too old, this would be a
very tight race that depends on knowledge of the node system
clock, but best to not have the race exist. Instead, we piggy
back off of the in flight tracking and do the check when
we start to handle the order, and commit at the end.
3. Change the functions that send orders and list unsent
orders to accept a time at which that operation is
happening. This way, in tests, we can pretend we're
listing or sending far into the future after the windows
are available to send, rather than exposing test functions
to modify internal state about the grace period to get
the desired effect. This brings tests closer to actual
usage in production.
4. Change the calculation for if an order is allowed to be
enqueued due to the grace period to just look at the
order creation time, rather than some computation involving
the window it will be in. In this way, you can easily
answer the question of "will this order be accepted?" by
asking "is it older than X?" where X is the grace period.
5. Increases the frequency we check to send up orders to once
every 5 minutes instead of once every hour because we already
have hour-long buffering due to the windows. This decreases
the maximum latency that an order will be reported back to
the satellite by 55 minutes.
Change-Id: Ie08b90d139d45ee89b82347e191a2f8db1b88036
* Add all new orders to the orders filestore instead of the database.
* Submit orders from the filestore to the new satellite SettleWindow
endpoint.
The orders filestore will eventually replace the orders DB completely.
For now, we will still be checking the orders DB and submitting those
orders if they exist. In a later release, we will completely remove the
orders DB, but we need both the DB and filestore for the transitionary
period.
Change-Id: Iac8780fd5ab770296181bbd313e1d335f072d4dc
on node's start checks if any of trusted satellites has GE status "Exited successfully"
if so - trying to delete blobs/satellite folder, so no trash left on SNO.
Change-Id: I566266c84f2a872df54cd01bc2f15a9934f138ed
Part 2 of moving usedserials in memory
* Drop usedserials table in storagenodedb
* Use in-memory usedserials store in place of db for order limit
verification
* Update order limit grace period to be only one hour - this means
uplinks must send their order limits to storagenodes within an hour of
receiving them
Change-Id: I37a0e1d2ca6cb80854a3ef495af2d1d1f92e9f03
See https://storjlabs.atlassian.net/browse/SM-752
These changes allow us to change the log level at runtime through a handler off of the debug endpoint.
Examples of changing the log level on storj-sim
To get the current level for the satellite api process:
curl -XGET 'http://127.0.0.1:10009/logging' --header 'Content-Type: text/plain'
To change the log level:
curl -XPUT 'http://127.0.0.1:10009/logging' --header 'Content-Type: text/plain' --data-raw '{"level":"error"}'
Change-Id: I05d164b290929fa06b6d78c01075ee41f8238044
Currently uploads can cause a lot of IOPS, reduce this by introducing a
in-memory buffer on-top of the file.
Change-Id: I5f4e3e01c0a36258271d180b922107de447bcb59
CreateTables hasn't been quite true for a while now, rename to
MigrateToLatest to be clearer in it's behavior.
Change-Id: Ida48e95122a5d9b7a814e922d3698e00024a2ba7
To improve delete performance, we want to process deletes asynchronously
once the message has been received from the satellite. This change makes
it so that storagenodes will send the delete request to a piece Deleter,
which will process a "best-effort" delete asynchronously and return a
success message to the satellite.
There is a configurable number of max delete workers and a max delete
queue size.
Change-Id: I016b68031f9065a9b09224f161b6783e18cf21e5
Currently storj-sim relies on the log lines to be exactly the same,
when they change it cannot find the necessary information from log.
Change-Id: Ia039915ef3375a7cf60f107b2c05c958de15b6d5
* debug
* traces
* cfgstruct
* process
Package `storj/private/version` will be removed as a separate change.
Change-Id: Iadc40faa782e6225513b28218952f02d9c240a9f
In many cases when a storagenode fails the preflight check, it is due to
test_table existing, which is used to determine read/write capabilities
after the initial schema verification. If preflight ends early due to a
failure or stopped storagenode, it may not get the chance to drop this
table.
This change excludes test_table from the schema comparison to ensure
that it never prevents a storagenode from starting up.
It also adds Preflight DB test for storagenode.
Change-Id: Ib8e71df2e42fda3b2a364fbf7a801891c5831d39
On satellite, remove all references to free_bandwidth column in nodes table.
On storage node, remove references to AllocatedBandwidth and MinimumBandwidth and mark as deprecated.
Protobuf message, NodeCapacity, is left intact for backwards compatibility.
Once this is released to all satellites, we can drop the column from the DB.
Change-Id: I2ff6c6537fc9008a0c5588e951afea58ede85838
When a storagenode begins to run low on capacity, we want to notify
the satellite before completely running out of space. To achieve this,
at the end of an upload request, the SN checks if its available space has
fallen below a certain threshold. If so, trigger a notification to the
satellites.
The new NotifyLowDisk method on the monitor chore is implemented using the
common/syn2.Cooldown type, which allows us to execute contact only once
within a given timeframe; avoiding hammering the satellites with requests.
This PR contains changes to the storagenode/contact package, namely moving
methods involving the actual satellite communication out of Chore and into
Service. This allows us to ping satellites from the monitor chore
Change-Id: I668455748cdc6741291b61130d8ef9feece86458
common/pb moved grpc to a separate package common/pb/pbgrpc.
This updates this repository to use it.
Change-Id: I2de2a190688871cf9cb61f7ea511f8a01e264e4e
Curently, storage nodes only report their capacity to satellites
once per hour. If a node fills up, it will fail all uploads until
the next contact cycle begins. With these changes, at the end of an
upload we check whether the MinimumDiskSpace threshold has been
passed. If so, trigger the monitor chore to update the node's
capacity, then trigger the contact chore to report the new
capacity to the satellites
Change-Id: Ie6aadaade1e2c12c87e03f8ff9059a50121380a0
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
Per default our server address is listening on all IP addresses on the machine.
This caused our preflight check to fail, as it did not have an hostname to lookup.
With this change, we are fine with this and go ahead.
Change-Id: I9eb5c891c099eb35f679d6d7e79ec38bb43b619f
Remove starting up messages from peers. We expect all of them to start,
if they don't, then they should return an error why they don't start.
The only informative message is when a service is disabled.
When doing initial database setup then each migration step isn't
informative, hence print only a single line with the final version.
Also use shorter log scopes.
Change-Id: Ic8b61411df2eeae2a36d600a0c2fbc97a84a5b93
- also updated ping chore to pick up trust changes
- fixed small typo in blueprint
- fixed flags for storj-sim
- wired up changes to testplanet
Change-Id: I02982f3a63a1b4150b82a009ee126b25ed51917d
* change satellite.Peer name to Core
* change to Core in testplanet
* missed a few places
* keep shared stuff in peer.go to stay consistent with storj/docs
libuplink was incorrectly setting timeouts to 10 seconds still, but
should have been at least 10 minutes. the order sender was setting them
to 1 hour. we don't want timeouts in uplink-side logic as it establishes
a minimum rate on tcp streams.
instead of all of this, just use tcp keep alive. tcp keep alive packets are
sent every 15 seconds and if the peer stops responding the connection
dies. this is enabled by default with go. this will kill tcp connections
when they stop working.
Change-Id: I3d7ad49f71950b3eb43044eedf4b17993116045b
all of the packages and tests work with both grpc and
drpc. we'll probably need to do some jenkins pipelines
to run the tests with drpc as well.
most of the changes are really due to a bit of cleanup
of the pkg/transport.Client api into an rpc.Dialer in
the spirit of a net.Dialer. now that we don't need
observers, we can pass around stateless configuration
to everything rather than stateful things that issue
observations. it also adds a DialAddressID for the
case where we don't have a pb.Node, but we do have an
address and want to assert some ID. this happened
pretty frequently, and now there's no more weird
contortions creating custom tls options, etc.
a lot of the other changes are being consistent/using
the abstractions in the rpc package to do rpc style
things like finding peer information, or checking
status codes.
Change-Id: Ief62875e21d80a21b3c56a5a37f45887679f9412
* storagenode/storagenodedb: Migrate to separate dbs
* storagenode/storagenodedb: Add migration to drop versions tables
* Put drop table statements into a transaction.
* Fix CI errors.
* Fix CI errors.
* Changes requested from PR feedback.
* storagenode/storagenodedb: fix tx commit
* test that all nodes can check in with all satellites
* keep kademlia config
* add untrusted satellite test
* use getversion
* remove kademlia config changes in test-sim-backwards.sh
* add kademlia flags back to storj-sim storagenode
* reset kademlia flags in storagenode entrypoint
What:
cmd/inspector/main.go: removes kad commands
internal/testplanet/planet.go: Waits for contact chore to finish
satellite/contact/nodesservice.go: creates an empty nodes service implementation
satellite/contact/service.go: implements Local and FetchInfo methods & adds external address config value
satellite/discovery/service.go: replaces kad.FetchInfo with contact.FetchInfo in Refresh() & removes Discover()
satellite/peer.go: sets up contact service and endpoints
storagenode/console/service.go: replaces nodeID with contact.Local()
storagenode/contact/chore.go: replaces routing table with contact service
storagenode/contact/nodesservice.go: creates empty implementation for ping and request info nodes service & implements RequestInfo method
storagenode/contact/service.go: creates a service to return the local node and update its own capacity
storagenode/monitor/monitor.go: uses contact service in place of routing table
storagenode/operator.go: moves operatorconfig from kad into its own setup
storagenode/peer.go: sets up contact service, chore, pingstats and endpoints
satellite/overlay/config.go: changes NodeSelectionConfig.OnlineWindow default to 4hr to allow for accurate repair selection
Removes kademlia setups in:
cmd/storagenode/main.go
cmd/storj-sim/network.go
internal/testplane/planet.go
internal/testplanet/satellite.go
internal/testplanet/storagenode.go
satellite/peer.go
scripts/test-sim-backwards.sh
scripts/testdata/satellite-config.yaml.lock
storagenode/inspector/inspector.go
storagenode/peer.go
storagenode/storagenodedb/database.go
Why: Replacing Kademlia
Please describe the tests:
• internal/testplanet/planet_test.go:
TestBasic: assert that the storagenode can check in with the satellite without any errors
TestContact: test that all nodes get inserted into both satellites' overlay cache during testplanet setup
• satellite/contact/contact_test.go:
TestFetchInfo: Tests that the FetchInfo method returns the correct info
• storagenode/contact/contact_test.go:
TestNodeInfoUpdated: tests that the contact chore updates the node information
TestRequestInfoEndpoint: tests that the Request info endpoint returns the correct info
Please describe the performance impact: Node discovery should be at least slightly more performant since each node connects directly to each satellite and no longer needs to wait for bootstrapping. It probably won't be faster in real time on start up since each node waits a random amount of time (less than 1 hr) to initialize its first connection (jitter).
this is a trivial operation for storagenode/console, as it doesn't
really need or use kademlia in the first place.
What:
Removes kademlia from storagenode/console
Why:
We are in the process of getting rid of kademlia, and this is one place where it's particularly easy.
Please describe the tests:
Existing tests exercise storagenode/console behavior; if they continue to work, everything here should be tested satisfactorily.
Please describe the performance impact:
None
* implement contact.checkin method
* add batching to update uptime checks
* rm batching
* rm other unneeded things
* fix lint
* fix unit test
* changes per CR comments
* couple more CR changes
* add identity check into grpcOpt
* fix lint
* why do you fix the test
* revert test change
* stop contact chore for repair test
* put node in cache
* comment out contact chore. See what happens
* Revert "comment out contact chore. See what happens"
This reverts commit 2e45008e36a50e0a842ae455ac83de77093d4daa.
* try stopping contact earlier
* stop contact chore in uplink_test
* replace self on chore with *RoutingTable for access to latest node info
* Revert "stop contact chore in uplink_test"
This reverts commit 302db70f4071112d1b9f7ee0279225ea12757723.
* Revert "try stopping contact earlier"
This reverts commit 806cc3b82f9d598899dafd83da9315a1cb0cb43c.
* Revert "stop contact chore for repair test"
This reverts commit dd34de1cfdfc09b972186c9ab9a4f1e822446b79.
* nicer flags
* fix concurrency
* add concurrent workers
* initialize things
* fix tests
* close retain service
* ensure we don't have workers working on the same satellite
* ensure things compile
* fix other compilation issues:
* concurrency changes
ran this with `go test -count=1000` and it passed all of them.
- we add a closed channel so that we can select on it with
context cancellation.
- we put a once in so we only close the channel once.
- every time the queue/running state changes, we have to broadcast
because we may want to wake up N pending Wait calls or other
concurrent workers.
- because we broadcast, we don't need to do the polling in Wait
anymore.
- ensure Run doesn't start multiple times so that we don't have
to worry about concurrent Close with multiple Runs.
- hold the lock while we start workers so that a concurrent Close
with Run can't decide that there's nothing started and exit
and then have Run start things.
- make sure to poll the closed/context channels through loops
or at the start of Run calls in case Close happens first.
- these polls should be under a mutex because they have a default
case which makes it possible to schedule such that Close hasn't
executed the channel close so it starts more work.
- cancel a local Run context when it's going to exit to make sure
that any retainPieces calls have a canceled context.
- hopefully enough comments to both check my work and help readers
digest what's going on.
Change-Id: Ida0e226a7e01e8ae64fa2c59dd5a84b04bccfbd7
* use the retain error class
Change-Id: I1511eaef135f98afd57b878e997e4c8a0d11cafc
* concurrency fixes again
- forgot to update the gc test to use the old Wait api.
- we need to drop the lock while we wait for the workers
to exit, because they may be blocked on the condition
variable
- additionally, we need to broadcast when we close the
signal channel because the state changed: they want
to wake up and exit.
Change-Id: I4204699792275260cd912f29aa73720f7d9b14b5
* undo my misguided rename
Change-Id: I6baffe1eb0434e260212c485bbcc01bed3250881
* remove pollInterval
* format paragraph more nicely
* move skew calculation into retain pieces
This PR introduces functionality for routine deletion of archived orders.
The user may specify an interval at which to run archive cleanup and a TTL for archived items. During each cleanup, all items that have reached the TTL are deleted
This archive cleanup job is combined with the order sender into a new combined orders service
Add retain service on storagenode. This service runs retain jobs that have been queued by the storagenodes. Rather than running retain jobs during the grpc Retain() call, the grpc call queues a retain job to the retain service and returns immediately afterwards, removing a significant bottleneck in garbage collection.
* add cache, update cache w/piece create/delete
* add service w/loop to cache to recalculate space used cache
* add piecestore cache to other sn svcs to use
* add table to persist the total space used
* rm cache where not needed
* rm stuff from sn svcs
* start fixing tests, changes per comments
* update commits
* add unit tests
* fix commiting before we write header bytes
* fix cache create test
* copy cache map, add started back to recalc
* fix test
* add test, update comments
Deprecate the pieceinfo database, and start storing piece info as a header to
piece files. Institute a "storage format version" concept allowing us to handle
pieces stored under multiple different types of storage. Add a piece_expirations
table which will still be used to track expiration times, so we can query it, but
which should be much smaller than the pieceinfo database would be for the
same number of pieces. (Only pieces with expiration times need to be stored in piece_expirations, and we don't need to store large byte blobs like the serialized
order limit, etc.) Use specialized names for accessing any functionality related
only to dealing with V0 pieces (e.g., `store.V0PieceInfo()`). Move SpaceUsed-
type functionality under the purview of the piece store. Add some generic
interfaces for traversing all blobs or all pieces. Add lots of tests.
* pkg/server: don't use global logger
* satellite/overlay: use correct logger
* pkg/kademlia: use correct logger
* linksharing: use conventional way to pass in logger
* use zaptest in tests
* re-organizing into bandwidth service. re-enable rollup loop
* Prevent uniqueness failure in bandwidth rollup
* Add test to make sure the rollup select date range works correctly
* add bandwidth config for rollup interval
