dgraph-io / dgraph-docs Goto Github PK
View Code? Open in Web Editor NEWA native GraphQL Database with a graph backend
Home Page: https://dgraph.io/docs
License: Other
A native GraphQL Database with a graph backend
Home Page: https://dgraph.io/docs
License: Other
Consolidate information on how to drop data.
Drop graphql data with mutation
Drop predicates, types ar all using /alter
https://dgraph.io/docs/deploy/single-host-setup/#run-dgraphs-ratel-ui
iirc dgraph-ratel
is not part of the dgraph binary right? So I think it does not make sense to still have this command here, especially since in the previous "Download Dgraph" section, we don't tell the readers about downloading Ratel.
Edit: Or we should follow the one at the later part of this page (https://dgraph.io/docs/deploy/single-host-setup/#run-dgraphs-ratel-ui-1), which is the second image.
The old page here: https://github.com/dgraph-io/dgraph-docs/blob/ed6f1dbb591234fa5c4eb704b593526b65a5050c/content/installation/kubernetes.md has a "removing a pod" section documenting how to use /removeNode and K8s delete to force a pod to be removed and rebuilt from a healthy pod with a correct p directory (rebuild in case of data issues on one node).
The removal process should be documented, perhaps in troubleshooting or how-to guides (or a link in troubleshooting about a bad p directory linked to the how to guide on removing an alpha)
Hey, oppening this issue just to let you know that there is a broken link on a "Next Page" button
The Count Results page's next button redirects to : https://dgraph.io/courses/datamodel/sql-to-dgraph/develop/read-data/paginate-results/
And it should redirect to: https://dgraph.io/learn/courses/datamodel/sql-to-dgraph/develop/read-data/paginate-results/
Dunno if oppening an issue here is the correct thing to do but just wanted to let u know
FAQ page here has not been updated in a long time. In particular:
We do not currently duplicate the various configuration flags reported by
dgraph alpha --help
and
dgraph zero --help
in the documentation.
All or key flags might be added here:
https://dgraph.io/docs/deploy/config/
We may simply point people to the "dgraph --help" output, but for find-ability and search indexing, at least the flag names should be in the docs.
Explain the relation between GraphQL and DQL :
https://dgraph.io/docs/deploy/fast-data-loading/bulk-loader/
The link (read some technical details
) here points to https://blog.dgraph.io/post/bulkloader/, which just redirects to https://dgraph.io/blog/ instead of the intended blog post (due to the redirection?). We should replace with https://dgraph.io/blog/post/bulkloader/.
The key generation command, when executed, gives gives an error and does not produce a key with the specified length.
Currently, documentation search only works for the docs' latest version (v21.12). I think the older versions are not indexed by algolia.
Any modification outside of these rules the PR will be refused.
1 - TODO -> Indentation of DQL, GraphQL and other code needs to be standardized(e.g: Tabs vs spaces - we need choose one and why). Because there can be no indentation conflict, if each one imposes their own way of working it becomes a mess. Each one will, for example, want to "repair" the previous indentation at their pleasure. It is not right.
2 - Modifications to current documentation content should have a clear and objective reason. If it's just a typo, it's valid without explanation. But any amendment that changes the concept must be rejected. In that case, please create something new to add in alignment with the context. Or explain how and why the change is needed.
cc @skrdgraph
Customer ran into this issue: https://discuss.dgraph.io/t/rfc-allow-language-tag-support-in-graphql/13027/12
The @dgraph directive for multi-lang predicates is not supported in v22 so that section should be removed.
we need an overview of DQL explaining the query language structure, the vocabulary (root, filter, func, etc ...) and the high level capabilities (filtering, recursivity, pagination etc ....) pointing to the language reference.
Align Dgraph overview page with new standard.
Describe what Dgraph is
Introduce personae for GraphQL and DQL and point to the rest of documentation accordingly
High level architecture "How it works"
Hosting options
Whats' next.|
In our docs we do not show quotes around superflags. However commands like
dgraph zero --audit output=audit-log-dir;compress=false;encrypt-file=enc-key
will fail to execute correctly without the quotes, because only the first flag will get parsed. In my PR I add quotes to a few commands related to audit logging. Presumably this is an issue elsewhere. I am not sure if it would be better to add double or single quotes.
Re-rewrite import data how to following structure used in #355 used for export data howto.
We favor relationship over Edges. The interesting info in thi section is just that we have directed relationship.
don't need to share internal representation of RDF IMO.
-> revamp necessary.
The next page of https://dgraph.io/docs/deploy/download/ should be "Configuration", but instead it is "Logging".
Consolidate and rewrite Migrate to Dgraph content.
https://dgraph.io/docs/deploy/single-host-setup/#run-using-docker-compose-on-single-aws-instance
https://dgraph.io/docs/deploy/multi-host-setup/
These two sections still uses the docker-machine
command, which is depecrated (docker/machine#4537) and replaced by Docker Desktop now. So, we should refactor these examples so that the readers are not confused.
DQL quickstart revamp.
Extract DQL quickstart (Rajie's work) from overview-351 branch.
https://dgraph.io/docs/deploy/dgraph-administration/#export-database
Here, it says that Since v21.03, the export and backup APIs are asynchronous
. But this is incorrect for export. Since v21.03.1
, export is synchronous again.
This is further verified by the fact that worker.Tasks.Enqueue()
only shows up once in the codebase, which is for backup only.
Consequently, we also need to refactor the Check queued tasks section a bit to not mention export.
The Dgraph releases pages needs a major revamp: https://dgraph.io/docs/releases/
In particular
This page: https://dgraph.io/docs/graphql/custom/custom-overview/ states that "Dgraph doesn’t execute your custom logic itself. It makes external HTTP requests" but I believe the more performant approach is to use DQL extensions which are executed within the Dgraph process.
I suggest
_"Dgraph creates a GraphQL API from nothing more than GraphQL types. That’s great, and gets you moving fast from an idea to a running app. However, at some point, as your app develops, you might want to customize the behavior to perform more general query logic and processing.
In Dgraph, you do that with code (in any language you like) that implements custom resolvers.
One way to do this is by creating custom DQL resolvers that run efficiently directly in the Dgraph process.
Alternatively, external processes can be called during the query or mutation process. Dgraph doesn’t execute that custom logic itself - it makes external HTTP requests. That means you can deploy your custom logic into the same Kubernetes cluster as your Dgraph instance, deploy and call, for example, AWS Lambda functions, or even make calls to existing HTTP and GraphQL endpoints."_
The back button in https://dgraph.io/docs/deploy/overview/ directs to the wrong (?) page. This page is the first page of this section (Deploy & Manage), but the back button redirects to "Ports Usage", which is a page somewhere further down in the same section. I'm not sure what's the intended behavior here though.
This command: "go get -u github.com/dgraph-io/ratel" at the link https://github.com/dgraph-io/ratel/blob/master/INSTRUCTIONS.md is not working.
"go get" seems to no longer be supported. The terminal output asks for a "go install" type of command instead.
Simplify GraphQL QuickStart and align with new doc templates
https://dgraph.io/docs/deploy/dgraph-administration/#restrict-mutation-operations
For the second command, correct me if i'm wrong but I don't think mutations-nquad=1000000
is relevant to the current context, as it is only demonstrating mutations=strict
. So I think we should remove it to not make readers confused.
There is also a missing quotation mark at the end of Enforce a strict schema by setting --limit "mutations=strict
.
When we add count index in our schema, the insertion of new tuples is very slow. Without count index we have around 50000 insertions/sec, but we add the count index, we insertions falls to 100 insertions/sec.
We think this is a bug, but we are not sure.
Thanks in advance
Create a HowTo entry for "importing and exporting your data"
Merge info from
https://dgraph.io/docs/cloud/admin/import-export/
https://dgraph.io/docs/deploy/fast-data-loading/overview/
https://dgraph.io/docs/deploy/fast-data-loading/live-loader/
https://dgraph.io/docs/deploy/fast-data-loading/bulk-loader/
and there is a reference to export import in https://dgraph.io/docs/cloud/cloud-api/backup/
Clean old pages and reference this howto when we mention export import.
In the Note at the very top of the encryption at rest, there are 3 links to other pages. For two of those:
Clicking on fast data load redirects you to the deploy and manage page with an option to then click on fast data load. This is inconsistent with most other links since the redirect directly to the page of the topic we click on. So, instead of it redirecting to the deploy and manage page, can we redirect it to the fast data load page?
Same with clicking on bulk loader. It redirects to a parent page with an option to then click on the bulk loader page. This is inconsistent with other links to the bulk loader page where they directly redirect to the bulk loader page.
https://dgraph.io/docs/deploy/dgraph-administration/
/admin/config/lru_mb
with /admin/config/cache_mb
(i verified from the commit history that the two of them has the same handler func)/admin/schema/validate
endpointWe need to document it on enterprise docs.
https://dgraph.io/docs/deploy/single-host-setup/#run-using-docker-compose-on-single-aws-instance
Right now, this part of using docker-compose to run dgraph on a single host is only mentioning AWS, but would it make more sense to not mention AWS, since docker-compose can be used anywhere.
This page does not add any info. to be removed.
Hi,
The above-mentioned page doesn't mention the following important things:
I think, there also should be information about what to do if many values bigger than 1 MB need to be stored.
For example: Can the threshold of 1 MB be increased?
Or: Can the vlog mechanism be deactivated? (I've seen your founder mention this possibility somewhere, but I believe, this wasn't implemented so far).
Edit:
Background/reference: A thread I started on the forum
It would also be useful to know if there is a built-in way to measure if the 1 MB threshold was reached. And also, what the 1 MB exactly is referring to (I'm assuming to all data of the key/value-pair).
Edit 2:
I'm also wondering if the 1 MB threshold is referring to the raw or the compressed data (being able to store 1 MB of compressed data vs. 1 MB of uncompressed data is a big difference).
If, after following the path in the title, you click on TLS Configuration Documentation, it leads you to a Get Started with Dgraph page, so it is not the right link.
Enabling ACL requires a secret key generation, but there is no command given with which they can generate said file. Also, in the example given below the directions, they used the following command:
echo '12345678901234567890123456789012' > hmac_secret_file
to generate the secret file. This is not randomly generated and is therefore not the best to use as an example.
Fixes: #305
https://dgraph.io/docs/deploy/download/#building-from-source
In here, the command given is make install
, but i found that running that gives an error (on an amd64 linux). But it is fine when I add the CGO_ENABLED=0
flag (something to do with jemalloc?).
So perhaps we should change it to CGO_ENABLED=0 make install
.
https://dgraph.io/docs/deploy/dgraph-administration/#secure-alter-operations
The two bottommost commands has the wrong port number, we should set it to the default which is 8080
.
https://dgraph.io/docs/deploy/dgraph-administration/#upgrade-database
The wal directory for zeros should be zw
, not w
Related to this PR dgraph-io/dgraph#8612
The docs need correspond changes.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.