Currently slower storagenodes can slow down deletion queue.
To make piece deletion faster reduce the maximum time spent in
either dialing or piece deletion requests.
With this change:
* dial timeout is 3s
* request timeout is 15s
* fail threshold is set to 10min
Similarly, we'll mark storage node as failed when the timeout occurs.
The timeout usually indicates that the storagenode is overwhelmed.
Garbage collection will ensure that the pieces get deleted eventually.
Change-Id: Iec5de699f5917905f5807140e2c3252088c6399b
table and drop not null constraint on objects column
Since, we want to move from charging our customers by object count to
segment count, this PR prepares the database to be able to record segments count
instead of objects count for satellite's billing system
Change-Id: Ie91ef354e78d24a268bc1cdc4327c182f733321e
This commit doesn't change any behavior, just organize the code in
different way to make it easier to implement different Criterias
to include nodes. Today we use NodeID and Subnet based selection
but later Criteria can be extended with different kind of
placement rules (like geofencing).
The change nodeselection is used by segment allocaton (upload) and repair
and excludes nodes from an in-memory selection.
Resolves https://github.com/storj/storj/issues/4240
Change-Id: I0c1955fe16a045e3b76d7e50b2e1f4575a7ff095
We needed Redundancy insided sat StreamID when uplink was defining RS values. Now it can be removed.
Change-Id: Id37187493eaa00cf29cb0262a050d71add3deb96
listPendingTransitionShim is a temporary transition shim intended to
make existing API processes keep working when a future DB schema change
is executed. For more explanation, see the message on commit c053bdbd70.
However, the shim has a small bug: it is missing the ORDER BY clause
that appears in the original ListPending method. This transition shim
code won't ever run until we make the DB schema change, so this bug
hasn't hurt anything yet; it's just important that we fix it before the
DB schema change happens.
Change-Id: I5953651583ee236500c2c07141dfc9d690a95118
This update is to set up users being able to register with a promo code added to their account in place of the free tier coupon.
Change-Id: I7badf87937b12664f145520b6dcc4b26fe750407
To make our free tier limits more clear, we will reduce the number of projects allowed from 3 to 1, and increase the storage and bandwidth limit of the free tier from 50 Gb to 150 GB. The total allotments across all projects for a given user are unchanged, just reduced to a single project.
Change-Id: Ic8dddb135f2b83a3f36e2b9fdcb477e351ec137b
We should improve the way how we are handling metabase errors in
metainfo endpoint.
https://storjlabs.atlassian.net/browse/PG-316
Change-Id: I1da6f333546cabf34d6eb1de8e94a3ef455d75d5
Move a variable declaration just before where is used for making easy to
follow the implementation when reading it.
Change-Id: I3e93597956352037a63c5496b11576fdd2c600a6
We don't need column uses_segment_transfer_queue in graceful_exit_progress
as now all exiting nodes are using graceful_exit_segment_transfer_queue and
table graceful_exit_transfer_queue has been dropped.
Change-Id: I4b7c087433f04138cf09bcf8ad3d8de2c185502a
The error message Stripe's API was sending:
"This property cannot be expanded (total_discount_amounts).
You may want to try expanding 'data.total_discount_amounts' instead."
Change-Id: I9f8cea4107d826d837755be2c3c04675a36f3c37
Multipart upload limits added. Last part has no size limit.
Max number of parts: 10000, min part size: 5 MiB
Change-Id: Ic2262ce25f989b34d92f662bde720d4c4d0dc93d
Speedup is done by reducing number of testplanet instances
for tests without changing main test logic.
Change-Id: Ic3849485d37b8ca55c013a45b7191dce65b88b04
Add some testing around adding/replacing promotional coupons and what
the user sees in the UI depending on their coupon status.
Change-Id: Ice6c0a0644d05af0c30c87a93ba963c0bb09e32d
Union query for both reputable and new nodes didn't properly work. The
top level query is required to have an `AS OF SYSTEM TIME` statement as
well.
Change-Id: I8ee6dd5b700c2b1ed2aa562962bfa72be7eec30a
Uplink needs only part of columns we are reading from DB.
To improve performance we should read only those that are
realy needed.
Change-Id: Ib39259318169c46afe5fa4c6ce2184da82e960c8
3b751a35c Removed our old coupon functionality, and slightly reworked
the invoice List() function in the stripecoinpayments package.
It turns out, this is causing some issues when trying to delete users.
This change keeps the new functionality, which is used in the satellite
UI, but under a new name, ListWithDiscounts()
Change-Id: I6a62a1de480e09d005dd22d75aa1e024fd2ed3a0
work
Since we have changed the repair worker to also mark a node as audit
failure if they return a not found error, we should ignore expired
segments when possible
Change-Id: Ie6a677e1d7b234e93965c736d05950440236653c
We don't use this column for anything. If you want to know if a node is
contained, you can check the pending_audits table.
Change-Id: I5671722a5fc6e1749d3a49e187a56556000ff941
We don't use this column for anything. If you want to know if a node is
contained, you can check the pending_audits table.
Change-Id: I8da1d8e01a2dcaff63c5067a7927b5451424ad04
this is so that multipart upload allows overwriting parts. our
consistency model is more relaxed because part overwrites
won't be atomic if they consist of multiple segments, but at
least this allows overwriting at all.
Change-Id: I21dac4c24046e05efe74e6c6fd189a02c95eb6c8
When an email is verified, insert an auth cookie so that when the user
is redirected after verifying their email, they are immediately taken to
the onboarding flow.
Change-Id: I557d8a2805b24dd8039ada255522bc1b56cc8b53
This change introduce problems with server side move so
let's revert it for now. Problem was found when latest
version of storj/storj was used in uplink tests.
This reverts commit 1ef06fae99.
Change-Id: I4d4fad5d1ea04ba15ff9d7bd765f7e078e9187c2
Removes database tables and functionality related to our custom
coupon implementation because it has been superseded by the Stripe
coupon and promo code system. Requires implementations of the
payments Invoices interface to return coupon usages along with
invoices.
Change-Id: Iac52d2ff64afca8cc4dbb2d1f20e6ad4b39ddfde
We were using mixed types for nonce fields. Protobuf
have storj.Nonce, metabase have []byte. This change
is a refactoring to have everywere its possible only
storj.Nonce.
Change-Id: Id54bd8481f30c721cdaf3df79206d25e7cfdab55
Cockroach doesn't like concurrent database creation, add a limiter to
avoid overloading the database.
Also limit the cockroach testing to the last 10 migrations.
Change-Id: Ie73c516ac2d362b251f605049e51eb25888432bd
We are not using the benchmark results for anything, they are mostly
there to ensure that we don't break the benchmarks. So we can disable
CockroachDB for them.
Similarly add short versions of other tests.
Also try to precompile test/benchmark code.
Change-Id: I60b501789f70c289af68c37a052778dc75ae2b69
* Add test cases to verify that all the endpoint that target a specific
entity respond 404 status code when the entity isn't found.
* Fix the endpoints that target a specific entity which responded a 500
status code response when the entity didn't exist to respond with 404
status code.
Additionally:
* Simplify some tests using an existing test helper function.
* Rename test functions to start with the entity name (e.g. Project,
User, etc.) for easing to run a set of test with the `-run` Go test
flag.
Change-Id: I82aad92e429207b72932ad4b79c08db6b486a19a
Populate the egress_dead column for taking into account allocated bandwidth that can be removed because orders have been sent by the storage nodes. The bandwidth not used in these orders can be allocated again.
Change-Id: I78c333a03945cd7330aec052edd3562ec671118e
It's safer to create new connection pool for piece deletion
only if dialer have no existing pool assigned.
Change-Id: I26661683ab7c0198587905478057c01c8f533a7e
We should be using object naming insted of path.
This is one place where we can easiliy change it.
To regenerate protbuf I had to remove gogo.proto.
Most probably it was confilicting with gogo.proto
from common/pb.
Change-Id: Ia5972f77994765c8f26bf1c3dc8205d2eadd70fa
We want to enable connection pool for piece deletion to avoid
doing multiple SSL hanshakes to SN while massive deletion process.
Change-Id: Ic917e4eda304ee16a286926ef046fe9e38bf38ca
We were manually converting ObjectKey fields to []byte to use it with
SQL query but we can just implement Value method to convert it automatically.
Change-Id: I6d346f4b59718e1e8ef37cd9f95e613b864b42cd
Update repair tests to check if audit score increases for nodes
that successfully send pieces during successfull and failed repairs.
Change-Id: Ie6abbde6155ab4697d209366c9fa497e731756e9
Drop table graceful_exit_transfer_queue which is not used anymore (replaced by graceful_exit_segment_transfer_queue).
Change-Id: Ie254fe9a54fb0784e350a439ce7a9bc99a3a58b5
We would like to verify if zombie object/segment works fine.
We need some metric for that. Figuring out number of deleted
objects is harder so let's for that later.
Change-Id: Ic99e2ce93256130b7c51f514824fddc009655075
Validates the new user data, through `Validate` method, on "create user"
endpoint earlier and remove specific checks of data that is already
validated by the `Validate` method.
Change-Id: I6b8f91c0b706f0c8316173c491e25c4858172640
A previous commit added a helper function for sending JSON data back to
the client.
This commit makes use of it for homogenizing the current implementation.
It also renames the existing helper message to send JSON errors to
starts with "send" because the new helper starts with it and they
helpers are clearer with their name starting with it.
Change-Id: I53ee0b4ca33d677a8ccd366c9ba6d73f4f472247
Fix the Admin API endpoints when it response with a client error
response code.
The most of the client response error send a JSON body but the
`Content-Type` header wasn't set to the appropriated value. This commit
fixes them and it adds assertions to the tests to very the
`Content-Type` header.
Updates the README to briefly document about the format of the client
errors endpoints responses.
Change-Id: Ifaf2122def801701211438ce241046be1adc0e8c
Move tests which where related with the `admin` package to the `console`
one because it's where the interface for the methods that the tests
check is defined.
Change-Id: I1bbccfe0badd989ed146a260af8d207d47737ef1
Migration tests are very heavy on database schema changes, which may
cause delays and retries. Separate out the migration tests and ensure
that they do not run concurrently on the same database.
Change-Id: I35b17525f18fd923546ce1fcc12d805c95073b6b
Why: big.Float is not an ideal type for dealing with monetary amounts,
because no matter how high the precision, some non-integer decimal
values can not be represented exactly in base-2 floating point. Also,
storing gob-encoded big.Float values in the database makes it very hard
to use those values in meaningful queries, making it difficult to do
any sort of analysis on billing.
Now that we have amounts represented using monetary.Amount, we can
simply store them in the database using integers (as given by the
.BaseUnits() method on monetary.Amount).
We should move toward storing the currency along with any monetary
amount, wherever we are storing amounts, because satellites might want
to deal with currencies other than STORJ and USD. Even better, it
becomes much clearer what currency each monetary value is _supposed_ to
be in (I had to dig through code to find that out for our current
monetary columns).
Deployment
----------
Getting rid of the big.Float columns will take multiple deployment
steps. There does not seem to be any way to make the change in a way
that lets existing queries continue to work on CockroachDB (it could be
done with rules and triggers and a stored procedure that knows how to
gob-decode big.Float objects, but CockroachDB doesn't have rules _or_
triggers _or_ stored procedures). Instead, in this first step, we make
no changes to the database schema, but add code that knows how to deal
with the planned changes to the schema when they are made in a future
"step 2" deployment. All functions that deal with the
coinbase_transactions table have been taught to recognize the "undefined
column" error, and when it is seen, to call a separate "transition shim"
function to accomplish the task. Once all the services are running this
code, and the step 2 deployment makes breaking changes to the schema,
any services that are still running and connected to the database will
keep working correctly because of the fallback code included here. The
step 2 deployment can be made without these transition shims included,
because it will apply the database schema changes before any of its code
runs.
Step 1:
No schema changes; just include code that recognizes the
"undefined column" error when dealing with the
coinbase_transactions or stripecoinpayments_tx_conversion_rates
tables, and if found, assumes that the column changes from Step
2 have already been made.
Step 2:
In coinbase_transactions:
* change the names of the 'amount' and 'received' columns to
'amount_gob' and 'received_gob' respectively
* add new 'amount_numeric' and 'received_numeric' columns with
INT8 type.
In stripecoinpayments_tx_conversion_rates:
* change the name of the 'rate' column to 'rate_gob'
* add new 'rate_numeric' column with NUMERIC(8, 8) type
Code reading from either of these tables must query both the X_gob
and X_numeric columns. If X_numeric is not null, its value should
be used; otherwise, the gob-encoded big.Float in X_gob should be
used. A chore might be included in this step that transitions values
from X_gob to X_numeric a few rows at a time.
Step 3:
Once all prod satellites have no values left in the _gob columns, we
can drop those columns and add NOT NULL constraints to the _numeric
columns.
Change-Id: Id6db304b404e6fde44f5a8c23cdaeeaaa2324f20
Why: big.Float is not an ideal type for dealing with monetary amounts,
because no matter how high the precision, some non-integer decimal
values can not be represented exactly in base-2 floating point. Also,
storing gob-encoded big.Float values in the database makes it very hard
to use those values in meaningful queries, making it difficult to do
any sort of analysis on billing.
For better accuracy, then, we can just represent monetary values as
integers (in whatever base units are appropriate for the currency). For
example, STORJ tokens or Bitcoins can not be split into pieces smaller
than 10^-8, so we can store amounts of STORJ or BTC with precision
simply by moving the decimal point 8 digits to the right. For USD values
(assuming we don't want to deal with fractional cents), we can move the
decimal point 2 digits to the right.
To make it easier and less error-prone to deal with the math involved, I
introduce here a new type, monetary.Amount, instances of which have an
associated value _and_ a currency.
Change-Id: I03395d52f0e2473cf301361f6033722b54640265
This PR utilize the new burst limit column from projects table to allow
control on the limit for request per seconds and token bucket size
When no burst limit is explicitly set, rate limit is applied to both so
we don't limit how quickly request can be made in a second.
Change-Id: I883235c60c5d6416aeadd1c80ed2ebd193aa4d9f
Currently loops wait for the coalesce duration for TriggerWait.
Let's skip the coalesce when we trigger it manually.
Change-Id: If5bacd4e263d233f1f3ea41b989922d2ed5a48d4
In order to limit the amount of overall requests a user can issue in a
time span, we need to have the ability to define such limit separate
from per second request rate.
This PR adds a new column on the projects table to store the burst limit
per project.
Change-Id: I7efc2ccdda4579252347cc6878cf846b85146dc7
Metrics observer is joining loop only for monitoring
and because of that it waits until non-monitoring
observer will join loop.
That was causing some delays in tests. To fix that
we are triggering non-monitoring observer manually.
Change-Id: Id93d59c9fd8d637dac0beea82ff2c40247d94b10
server-side move extended with moving between buckets, for this reason
we change bucket name for object in db.
Change-Id: Ie21bcccc170e6ff14dcd8053fdb86fdf6d8438a0
Some processing inside storagenodes is async compared to uplink upload
and download, hence we need to explicitly wait for storagenodes to
finish their pending work before flushing orders to the satellite.
Hopefully this fixes TestAttributionReport flakiness.
Change-Id: I77c651ab6471ae094b5c21d1ab3860c96cb0d039
Not all errors from RunOnce can be retried. The context can be cancelled
with several different errors, e.g. timeout. Ensure we stop the loop
when context has errored, because none of the queries will succeed
when it has failed.
Change-Id: If3ff11f11a6f43c0d67633be1cfaf23e3e9e55f3
Second method needed to perform server-side move. It updates
metadata key and nonce and all segments key and nonces.
Change-Id: Ia43b26622a13048269f0ae9e1524b345db112adb
Since we are sharing the reporting logic between repair and audit. We
need to remove metric reporting logic in reporter.
Change-Id: Ib87295ab19079329e7438327d785a7f5c21d3b21
In the situation where the flushes take longer than the incoming
rate of writes, the RollupsWriteCache will take every connection
in the database pool and use them forever. Instead of doing that
and taking down satellite availability, bound the number of flush
operations that it will perform and drop incoming writes earlier
to keep memory usage constant.
Adds monitoring events for if any flushes or updates are lost.
Change-Id: I81b169b73501ee9b999f4b03d1e79645fc56f167
First from two methods needed to perform server-side move. It gets
metadata key and nonce and all segments key and nonces and returns
all of that to uplink.
Change-Id: Ied2c79559e77d3f63091c4d61948f2d6a2147d67
Remove the logic associated to the old transfer queue.
A new transfer queue (gracefulexit_segment_transfer_queue) has been created for migration to segmentLoop.
Transfers from the old queue were not moved to the new queue.
Instead, it was still used for nodes which have initiated graceful exit before migration.
There is no such node left, so we can remove all this logic.
In a next step, we will drop the table.
Change-Id: I3aa9bc29b76065d34b57a73f6e9c9d0297587f54
Currently TextMaxVerifyCount flakes in some tests, try increasing the
sleep time to ensure that things are slow enough to trigger the error
condition.
Also pass ctx to all the funcs so we can handle sleep better.
Change-Id: I605b6ea8b14a0a66d81a605ce3251f57a1669c00
Currently, requests that were successfully passed through the metainfo
endpoints rate-limiter might still fail in the middle of the
corresponding response. The problem is that we perform rate-limiting a
second time, which means other requests would influence whether the
current (already rate-checked) request will fail. This also has other
unintended effects, like responding with rpcstatus.PermissionDenied for
requests that were successfully rate-checked and did not lack
permissions but were rate-checked again in the middle of
(*Endpoint).BeginObject. This situation has been happening on the
gateway side and might affect other uplink clients. This change, where
appropriate, swaps subsequent validateAuth with validateAuthN that
performs rate-limiting once.
Change-Id: I6fc26dedb8c442dd20acaab5942f751279020b08
Current default is equal to hardcoded maximum of batch size
for loop listing. With this change we will bump maximum
batch size but current default won't change. We will be able to
increase it later without need for source code change.
Change-Id: I2744a87be28af4157f58ede73455682f61733bc1
We were using this method for metainfo loop but now
segment loop is not using it so we can drop it.
Change-Id: I60c5b4f86a619259906d8c2ba76e665b8715be75
At some point we moved metabase package outside Metainfo
but we didn't do that for satellite structure. This change
refactors only tests.
When uplink will be adjusted we can remove old entries in
Metainfo struct.
Change-Id: I2b66ed29f539b0ec0f490cad42c72840e0351bcb
* Disabled updating project for paid tier users who have 0 storage or
bandwidth limits set (these users have been disabled and should not be
able to increase their limits through the UI)
* Better error handling
* Removed unnecessary type conversions
Change-Id: I1e07a1909a4ba877718aa944848f36382f7dbbe2
GetRandomStripe function to randomly select a segment stripe to
audit was using `segment.EncryptedSize/segment.Redundancy.StripeSize()`.
Since integer divsion truncates it leads to skipping last stripe if
its size is less than stripe size. Use `Redundancy.StripeCount` to
get correct stripe count.
Change-Id: Ida09e035be30a21219ab3e1aedd66af8be707d1b
Don't update the project description if the request body has the
description field set to an empty string.
This follows the same convention used for updating an user's account.
Change-Id: I027047e609760e033cf4b233b1be352c6bf0ec8f
the month
The Stripe API had a bug before that it wasn't calcualting the input
timestamp based on correct timezone. We had a workaround to not include
the last day of the month in our code when submitting to Stripe.
Now, Stripe has fixed the issue. We need to remove the workaround and
include the last day of the month into our invoice generation
Change-Id: Ic6364ed071be73a19f0b0b46f274a02fb2489db5
