Commit Graph

77 Commits

Author SHA1 Message Date
Michal Niewrzal
03f8ad323d satellite/metabase: remove segment_copies support from ListSegments
We don't need to support segment copies with references anymore.
We migrated to copies where all metadata are copied from original
segment to copy.

https://github.com/storj/storj/issues/5891

Change-Id: Ic91dc21b0386ddf5c51aea45530024cd463e8ba9
2023-08-08 11:21:08 +00:00
dlamarmorgan
0f4371e84c scripts/tests/{backwardcompatibility,integrations}: add test scripts
Change-Id: Ib83cd0f083bab7f560a200fd95e62e5b21e60c27
2023-07-17 18:39:18 +00:00
paul cannon
97e20bc579 scripts/tests: fix rollingupgrade test even more
This might be pretty awful, but at least it is a complete and non-flaky
solution.

**Only when using the rollingupgrade test** (which implies a throwaway
satellite and also a PostgreSQL backend), create a trigger on the nodes
table which forces last_net to be equal to last_ip_port always.

Change-Id: I8448cf131e46576d96a414d06780270c7b2b1892
2023-03-13 15:49:07 +00:00
paul cannon
fd6ce6b9a5 scripts/tests: fix test-sim-rolling-upgrade.sh
This test involves a satellite with dev defaults (DistinctIP=no) being
upgraded past commit 2522ff09b6, which
means we need to run the dev-defaults-satellite-upgrade migration SQL
to avoid getting DistinctIP=yes behavior (which breaks the tests).

Change-Id: I29fb596d1ffa568dad635d98cfe9abacd3aaa48f
2023-03-09 23:35:36 +00:00
paul cannon
2522ff09b6 satellite/overlay: configurable meaning of last_net
Up to now, we have been implementing the DistinctIP preference with code
in two places:

 1. On check-in, the last_net is determined by taking the /24 or /64
    (in ResolveIPAndNetwork()) and we store it with the node record.
 2. On node selection, a preference parameter defines whether to return
    results that are distinct on last_net.

It can be observed that we have never yet had the need to switch from
DistinctIP to !DistinctIP, or from !DistinctIP to DistinctIP, on the
same satellite, and we will probably never need to do so in an automated
way. It can also be observed that this arrangement makes tests more
complicated, because we often have to arrange for test nodes to have IP
addresses in different /24 networks (a particular pain on macOS).

Those two considerations, plus some pending work on the repair framework
that will make repair take last_net into consideration, motivate this
change.

With this change, in the #2 place, we will _always_ return results that
are distinct on last_net. We implement the DistinctIP preference, then,
by making the #1 place (ResolveIPAndNetwork()) more flexible. When
DistinctIP is enabled, last_net will be calculated as it was before. But
when DistinctIP is _off_, last_net can be the same as address (IP and
port). That will effectively implement !DistinctIP because every
record will have a distinct last_net already.

As a side effect, this flexibility will allow us to change the rules
about last_net construction arbitrarily. We can do tests where last_net
is set to the source IP, or to a /30 prefix, or a /16 prefix, etc., and
be able to exercise the production logic without requiring a virtual
network bridge.

This change should be safe to make without any migration code, because
all known production satellite deployments use DistinctIP, and the
associated last_net values will not change for them. They will only
change for satellites with !DistinctIP, which are mostly test
deployments that can be recreated trivially. For those satellites which
are both permanent and !DistinctIP, node selection will suddenly start
acting as though DistinctIP is enabled, until the operator runs a single
SQL update "UPDATE nodes SET last_net = last_ip_port". That can be done
either before or after deploying software with this change.

I also assert that this will not hurt performance for production
deployments. It's true that adding the distinct requirement to node
selection makes things a little slower, but the distinct requirement is
already present for all production deployments, and they will see no
change.

Refs: https://github.com/storj/storj/issues/5391
Change-Id: I0e7e92498c3da768df5b4d5fb213dcd2d4862924
2023-03-09 02:20:12 +00:00
Sembeth
0408997e6c
cmd/multinode: Removes dependency on deprecated identity-dir flag, code and documentation. (#5646)
* The Multinode Dashboard no longer requires the identity files.

* Removed deprecated code that used the identity-dir.
2023-03-08 13:56:15 +01:00
Michal Niewrzal
8aa6110835 tests/backwardcompatibility: use noquic tag for release binaries
Older releases are not compiling with latest Go version if quic is used.
We need to add noquic tag to be able to compile older release with
latest Go version.

Change-Id: Id5768fcaa5c1f7cf3e6fbb633e7ca60309b7a37c
2023-02-21 13:26:46 +01:00
Michal Niewrzal
d21bbab2b2 satellite: fix metabase configuration wiring
New flag 'MultipleVersions' was not correctly passed from metainfo
configuration to metabase configuration. Because configuration was
set correctly for unit tests we didn't catch it and issue was found
while testing on QA satellite.

This change reduce number of places where new metabase flags needs
to be propagated from metainfo configuration to avoid problems with
setting new flags in the future.

Fixes https://github.com/storj/storj/issues/5274

Change-Id: I74bc122649febefd87f665be2fba628f6bfd9044
2022-11-02 15:17:34 +00:00
Egon Elbre
947cdf0fb4 ci: fixes for new Go version
We need to make exceptions for older uplink versions, because it does
not compile with newer Go versions due to quic dependency.

Change-Id: I3e073694f0942029c56740f0689088058ee068c3
2022-11-02 15:17:46 +02:00
dlamarmorgan
d39caf7674 cmd/satellite: change invoicing flags to iso8601
Update input parameters for invoicing commands to use iso8601 standard formatting.

Change-Id: I3b4a7df268e8f1a5cd51b0dab03ed38e13a76554
2022-10-07 07:58:23 -07:00
Michal Niewrzal
a3d9630336 scripts: move tests to separate folders
Doing some cleanup in "scripts" folder. All integration like tests are
moved under "test" directory (integration, bc, redis) and bash scripts
are adjusted to reflect new location.

As an addition "scripts/install-awscli.sh" was deleted as it was not
used.

Change-Id: I152905c4258f471a71f2d0e8731d91bb075e99c1
2022-09-16 11:10:18 +00:00
JT Olio
098ee71e03 test-sim-versions: write what bash step is executing
Change-Id: If1231f69601c1774bc6ce116f45105cc1b2f4e22
2022-06-10 19:33:03 +00:00
Jeff Wendling
ccb847a329 scripts/tests: write old uplink config yaml directly
sometimes these scripts want to have an access imported
after it has been potentially modified by having the
satellite address and node id added. it used to use the
uplink command to do this, but the cli api for that
has changed. rather than try to have the script detect
which uplink version is in use and call the right thing
it can always write out a valid yaml file and depend on
the new cli migrating it.

Change-Id: Ib82819699333f5f29e00117b99bfb10640033b94
2022-02-22 15:26:06 -05:00
Jeff Wendling
863b01bf09 testversions: skip some downloads for older uplinks
uplink command versions >= 1.48.0 always do multipart
uploads which cannot be downloaded by any of the
versions in the test < v1.27.6. so skip those tests.

Change-Id: I9644afbd14bfce9facfd87644d132f7d66367d62
2022-02-22 17:24:07 +00:00
Jeff Wendling
0a860a2e89 scripts: fix builds for new uplink command
the new uplink command expects to be able to migrate
and so we need to specify the --legacy-config-dir
flag sometimes as well. but unfortunately, the old
uplink will error if it gets a flag it doesn't
understand, so we have to set it as an env var.
but we can't use the env var for the --config-dir
flag all the time because the old uplink doesn't
look for that env var.

Change-Id: I019315192c0e6c348814527794342d823a5f9ec3
2022-02-16 21:40:06 +00:00
Artur M. Wolff
785eb93cd7 ci: install gateway@latest instead of @main
Gateway-ST frequent release cycle has been resurrected, which means it's
safer to use the latest release tag in the storj repository's CI now.

Change-Id: I9df1c789a9b9418ba7cceaec9cfec3cc6c448284
2021-11-01 12:16:38 +00:00
Qweder93
50baefa10e satellite/metabase: limit maximum number of parts and size
Multipart upload limits added. Last part has no size limit.
Max number of parts: 10000, min part size: 5 MiB

Change-Id: Ic2262ce25f989b34d92f662bde720d4c4d0dc93d
2021-10-22 10:16:37 +00:00
igor gaidaienko
5b4a9070b0 scripts/testversions: reduced the number of version to test 2021-10-12 13:28:46 +03:00
Artur M. Wolff
cc9b845a83 ci: switch back to pulling gateway@main
Switch back to pulling gateway@main for continuous integration because
https://review.dev.storj.io/c/storj/gateway/+/5834 was merged. This
change is a partial revert of
https://review.dev.storj.io/c/storj/storj/+/5832.

Change-Id: I0a87f6f18a9a863fe92003b76c830335f9253ced
2021-09-28 15:01:04 +00:00
Egon Elbre
8ef03b0967 ci: cleanup ws before build, fix gateway install
Jenkins uses the same folder for different PR-s. Sometimes other runs do
not cleanup after themselves (e.g. timeout), hence add a cleanWs step to
ensure we delete files in the workspace.

gateway-st introduced a replace directive in go.mod, which does not work
with go install. Hardcode to the last version without the directive.
Using this fix to unblock ci builds.

Change-Id: I5e5d75bf47e30a5a8b6d835867c0c9176f25e08a
2021-09-23 15:57:47 +03:00
Yaroslav Vorobiov
b426248936 scripts/tests: fix missing multinode binary
Change-Id: I288d06c9eaa814ddca1e4aaacd86c5277aba86cc
2021-05-14 22:47:05 +00:00
Kaloyan Raev
a264a4422b cmd: remove metainfo-migrator and metainfo-verifier
We already merged the multipart-upload branch to main. These two tools
make only sense if we are migrating a satellite from Pointer DB to
Metabase. There is one remaining satellite to migrate, but these tools
should be used from the respective release branch instead of from main.

Removing these tools from main will:
1) Avoid the mistake to use them from the main branch instead of from
the respective release branch.
2) Allow to finally remove any code related to the old Pointer DB.

Change-Id: Ied66098c5d0b8fefeb5d6e92b5e0ef5c6603df5d
2021-04-08 14:50:29 +00:00
Michał Niewrzał
a5bc38b197 build: replace 'multipart-upload' branch occurences
We removed `multipart-upload` branches but we have few leftovers.

Change-Id: I74f637748a248cb286cd190a3a8f895ef85f3940
2021-03-31 11:06:47 +02:00
Michał Niewrzał
9bc2a9a4b7 rollingupgrade: change how metainfo-migration tool is installed
Change-Id: Iad1285a770f76c9cf09a872a4bf1cd39969ee117
2021-03-26 15:03:39 +00:00
Moby von Briesen
2a68e0767e scripts/tests/testversions: Fix testversions Episode 2
The previous fix did not remove the comment at the front of the config
values that were modified.

Change-Id: I05729c30a5398d049c6254afec01df00230d8deb
2021-03-25 18:06:13 -04:00
Moby von Briesen
63cd807c9f scripts/tests/testversions: Increase bandwidth/storage limits for testversions
A recent change made the default usage/storage limits for projects 50gb
rather than 500gb. This increases the default limit for testversions.

Change-Id: Ibea05c0d0760662e447b6455d560a2a640801c6c
2021-03-25 14:43:28 -04:00
Michał Niewrzał
0ef0a3f967 rollingupgrade: add migration to metabase
For time of transition from pointerdb to metabase we need add migration step to rollingupgrade tests and comment few cases.

Change-Id: Ib12ae6aa14be35f9bf4ff3efb55cfc6957d4ceba
2021-03-25 10:52:35 +00:00
Michał Niewrzał
908a96ae30 Merge remote-tracking branch 'origin/main' into multipart-upload
Change-Id: I075aaff42ca3f5dc538356cedfccd5939c75e791
2021-02-11 11:48:23 +01:00
Moby von Briesen
520da5b1e4 scripts/tests/rollingupgrade: fix rollingupgrade test
Fix logic that determines whether satellite has the marketing web server
based on version number.

Change-Id: I34df745d16acf1ce1478fe992448b18565e6d363
2021-02-08 20:06:00 -05:00
Yingrong Zhao
babe01f229 scripts/tests/rollingupgrade: fix rollingupgrade test backwards
compatibility

Change-Id: I1551f993bc8ca5e6fac07da7710dd46dbb031e47
2021-02-08 21:37:24 +00:00
Michał Niewrzał
ad3e3a38c5 Merge 'main' branch
Change-Id: Ia0db1b1f9ef3e0671d3f2208881b0abc3064e200
2021-01-04 12:13:45 +01:00
Stefan Benten
7f1871b8f1 all: switch from master to main 2020-12-28 22:59:06 +01:00
Michal Niewrzal
218bbeaffa Merge 'master' branch
Change-Id: Ica5c25607a951076dd9f77e35e308062f71ce3f0
2020-12-07 15:05:52 +01:00
Yingrong Zhao
746315672f scripts/tests/testversions: fix indentation
Change-Id: Iaa5aec27f0ad78e1d8bf1a68aa5a62762c8ab537
2020-12-04 21:54:55 +00:00
Yingrong Zhao
0faf7d5293 scripts/tests/testversions: fix race in install_sim
Change-Id: I0792686d99a222d5977fd913425e2b94d100c40e
2020-12-04 18:18:14 +00:00
Kaloyan Raev
cc9e9ee1f5 storj-sim: use gateway from multipart-upload branch
Change-Id: I0886d277b3b757c8b00975a3e95c2d0d1228488b
2020-12-04 15:15:47 +02:00
Yingrong Zhao
13555f3983 scripts/tests/testversions: enable concurrent installation for each
version

The test-versions test currently takes 1h 40min to run each time. By
running each installation concurrently, hopefully, it will reduce the execution
time for the whole test.

Change-Id: I680c7d9945e982894b11825c9075c167f754e087
2020-12-03 15:01:37 +00:00
Egon Elbre
e19fabc880 scripts/tests/rollingupgrade: fix typo in flag
Change-Id: Ia3e8d076741a30fb2a42af9b2621796a814c75ae
2020-11-18 19:15:42 +00:00
Egon Elbre
8da5e6a554 scripts/tests/rollingupgrade: use wait-for instead of sleep
Change-Id: Ie879e061d3b312705726375953767d420e922073
2020-11-18 12:00:16 +00:00
littleskunk
9ab824d3e6
jenkins/rollingupgrade: sleep 5 seconds between old api startup and database migration (#3971) 2020-11-16 21:25:11 +01:00
Yingrong Zhao
54c5d564a1 scripts/tests/testversions: fix older uplink setup
This PR does follwing changes:
    1. Change oldest uplink version in the test to v0.35.3
        When the test is first created, we decided to support uplink
        version starting from v0.17.1, however with many API changes,
        older uplinks are not usable with latest version of the network
        anymore. One of the reasons being older uplinks uses deprecated
        endpoint. Therefore, we will change the oldest uplink version to
        the one that's using only new endpoints.
    2. Disable tls certificate verification in uplink
    3. Use storj-sim version control server instead of production one
    4. Skip uplink version v1.3.x due to bug in that release

Change-Id: I926a6bb9829cb7181ee752437cdcb67e59197fe0
2020-11-11 17:00:01 -05:00
Yingrong Zhao
8fd841b910 scripts/tests/testversions: fix installation during setup
This PR fixes below issues:
1. remove concurrent installation for various versions
    We were doing this to decrease the amount of execution time the versions test.
    However, it's returning incorrect exit code when there's an
    installation failure.
    Right now, we are only installing two versions of `storj-sim` and
    the rest are only doing uplink cli installation. The performance of
    this test should be hugely impacted by the setup step now.
2. only remove release settings instead of deleting the entire file
    uplink CLI is referrencing `private/version` package. Therefore, we
    cannot delete it
3. add back `GATEWAY_0_API_KEY` in storj-sim
    In order to set up older version of uplink cli, we need access to
    the gate way api key.

Change-Id: Ia3c37c197bd007b6e1f7c2bd71adde42181d46f0
2020-11-10 20:38:49 +00:00
Jessica Grebenschikov
99c88efbbf scripts/tests: fix gateway tests
Change-Id: I9a23ef08794043ad615066ae5929df9ff3a02d69
2020-10-27 08:21:28 -07:00
Yingrong Zhao
746cbfc659 scripts/tests/rollingupgrade: test current release version on master
branch

Currently, we are testing previous release version upgrading to latest
master on each master build
However, this behavior is only desired when the test is running on a
release branch.

Change-Id: Iaeb66f44951c9e4934ca3c8316d1e490d7958239
2020-10-22 11:45:54 -04:00
Kaloyan Raev
bd177bff03 cmd/storj-sim: cleanup gateway setup
Remove usage of --non-interactive flag. It is not provided (and
necessary) by the multitenant S3 gateway anymore.

ACCESS_KEY and SECRET_KEY env vars are not provided anymore as they are
not generated by the multitenant S3 gateway.

Change-Id: I3ecfb92110e31ae13977de3899dad273daae6c1e
2020-10-06 14:22:47 +00:00
Ethan Adams
66f5368807
satellite/testing: Change testing to use PG 12.3 (#3913)
Co-authored-by: Stefan Benten <mail@stefan-benten.de>
2020-06-25 20:17:39 +03:00
Stefan Benten
db57e5cf7b
scripts/tests/rollingupgrade: continue for v1.6.4 (#3912) 2020-06-25 13:57:39 +02:00
Yingrong Zhao
c78f88beb6 scripts/tests: use the same version for storj-sim as satellites
Change-Id: If85b14a0a5af334b9547d73bc4dbe65cdd856984
2020-06-05 13:17:17 -04:00
Yingrong Zhao
175f048aa7 scripts;cmd/storj-sim: include satellite node id in satellite address
Expose satellite ID from storj-sim so we can have access to it when
changing satellite address.

Change-Id: Ife816f4d35eae2d0bc9f9ad592fe75d73d93d9ff
2020-06-05 09:44:19 -04:00
Yingrong Zhao
9a04ca0527 scripts/tests/rollingupgrade: skip v1.6.3
We removed the ability to change satellite address in for uplink cli v1.6.3. Therefore we will skip that version for now

Change-Id: I0f78cfe27b51fd6cad571636ba38266f5f672d58
2020-06-05 09:00:36 +00:00