From the name of the function and from the way it is used (only called
in one place, from "storj.io/storagenode/gracefulexit".(*Chore).Run()),
it should not return graceful exits that have already completed.
In particular, this causes a problem in the case that a node has already
completed a graceful exit from one satellite, after which the satellite
was decommissioned and no longer in the "trusted" list. This causes an
error message to show up in the node logs every single minute like
"failed to get satellite address ... satellite \"X\" is untrusted".
https://forum.storj.io/t/error-gracefulexit-service-failed-to-get-satellite-address/11372
This change causes ListPendingExits to list pending exits only, not all
exits.
Correspondingly, the check for whether an exit is already completed, in
(*Chore).Run(), becomes unnecessary and is here removed.
Change-Id: Ia3e9bb3e92be4a32ebcbda0321e3fe61d77deaa8
The rollup archiver chore moves bucket bandwidth rollups and
storagenode rollups that are older than a given duration
to two new archive tables.
Change-Id: I1626a3742ad4271bc744fbcefa6355a29d49c6a5
When using calling time.Now() multiple times, they can cross
month boundary causing errors in calculations.
Change-Id: I66b5be7598f3bf475b4b5fe0dcce82eee55b3134
Full scope:
storagenode/{console,nodestats,notifications,reputation,storagenodedb},
web/storagenode
These columns are deprecated. They used to be for the uptime reputation
system which has been replaced by downtime tracking with audits.
Change-Id: I151d6569577d89733ac97af21a1d885323522b21
This index is obsolete and duplicates a similiar (project_id, name)
index on the same table.
Moreover, it might confuse CockroachDB which of the two index to use,
which may might affect DB performance.
Change-Id: If8d1df8347714942cea9dca82864ba5f4973bed3
This PR introduces a new listener that can listen for quic traffic on
both storagenodes and satellites.
Change-Id: I5eb5bc82c37dde20d3be2ec8fa5f69c18fae0af0
Provide a clearer error message to users who confuse the API Key with
the Access Grant and suggest the right command to them.
Change-Id: If73ae8cde140b68a19f4cfc3f59bb88a3b74c9c1
When we observed the value for total piecesizes stored in the network,
we were doing it after converting them to byte-hours, rather than using
the actual piece sizes. This fixes that issue.
Change-Id: I1564d21b519f70eb59f298d97dbd777baf127723
We wanto have single uplink branch for standard and multipart-upload satellite but some tests are using helper methods from multipart. This change adds methods used by uplink test.
Change-Id: I82352ed56674ff7e8743b58061ba594018e78e3b
On servers with non-UTC it would have calculated a different month boundary.
If node joined in current month calculations will be related on amount of days node've been working.
Change-Id: Ie572b197f50c6cdff5a044a53dfb5b9138f82f24
Create a storj-sim test that checks that uplinks operations works when
satellite runs and can connect to Redis and when it cannot connect to
simulate a Redis downtime. Also verifies that the satellite can start
despite of Redis being downtime.
This test currently doesn't pass and it will be the one used to verify
the work that has to be done to make sure that the satellite allow the
clients to perform their operations despite of Redis being unavailable.
We require these changes before we deploy any customer face satellite on
a multi-region architecture.
NOTE that this test will be added later on to Jenkins to run this test
every time that we apply changes and at that time we'll see if it has to
be adjusted for being able to run on Jenkins because as it's now it may
not work because the scripts start and stop a Redis docker container.
Change-Id: I22acb22f0ca594583e36b45c88f8c03bac73b329
Currently first satellite GC would've conflicted with second satellites
public RPC port. Instead assign "satellite workers" a new peer class.
Change-Id: Id6bdaa17243556482e88da708c5147149788f6be
Due to the issues with the licensing and go version requirement by the
quic-go library, we can not introduce the library into libuplink at the
moment. Therefore, we decided to put our quic integration code into
storj/storj. It will unblock us from rolling out quic to storagenodes.
Change-Id: If9d765da10c45947a4e3a3a11e4679bab69bcf08
WHAT:
add a banner with a message saying: gateway MT is in beta for EU and US2 sats
WHY:
user experience?
Change-Id: I0964c4499cdaf12dbd9dba082910fff4ff6f9a12
* web/satellite: Pin Project Owner in Team tab to top of members list
* web/satellite:Pin Project Owner to top of members list
- input missing semicolon
* web/satellite: Pin Project Owner to top of members list
- refactor project owner variable instantion
- check for existance of owner before returning members list
* web/satellite: Pin project owner to top of members list
- changed returned members list variable name
- add comment explaining unshift logic
Add ProjectsCursor type for pagination
Add PageCount, CurrentPage, and TotalCount ProjectsPage
This allows us to mimic the logic of GetBucketTotals and the
implementation of BucketUsages in graphql for the new ProjectsByOwnerID
functionality.
Change-Id: I4e1613859085db65971b44fcacd9813d9ddad8eb
We add a timeout for the http client used to register the access with
the auth service. We have a hard-coded common default for now.
Change-Id: I50207ad83c9221b7cb61f39310e24b140b95673b
Full scope:
private/testplanet,satellite/{overlay,satellitedb}
Description:
In most cases, downtime tracking with audits will eventually lead
to DQ for nodes who are unresponsive. However, if a stray node has no
pieces, it will not be audited and will thus never be disqualified.
This chore will check for nodes who have not successfully been contacted
in some set time and DQ them.
There are some new flags for toggling DQ of stray nodes and the timeframes
for running the chore and how long nodes can go without contact.
Change-Id: Ic9d41fdbf214736798925e728245180fb3c55615
Allow the satellite commands which uses the live accounting cache (core
and API) to run when at the time that its instantiated there is an error
connecting to the backend.
This prevent that if live accounting backend is down we can run these
services because:
1. The services must run despite of the cache backend being down
although it may be degraded.
2. We may need to start new replicas of the services or the services in
a different place while we are troubleshooting and fixing the cache
backend system.
3. Our services may restart when the cache backend or the network
connecting to it fails momentarily.
Change-Id: Ic93f9571bc0865c9488d64ab1356376fae797efc
Tally shouldn't abort its cycle if the accounting cache return an error
because it isn't an essential requirement to update it.
Change-Id: I78fd2bd9cf253ddedfb9ada80c0fa2ddf438f647
