ref
https://linear.app/ghost/issue/ENG-1592/start-monitoring-connection-pool-utilization-in-ghost
- This commit adds prometheus metrics to the connection pool so we can
start to track connection pool utilization, number of pending acquires,
and also adds some basic SQL query summary metrics like queries per
minute and query duration percentiles.
- The connection pool has now been theorized to be a main constraint of
Ghost for some time, but it's been challenging to get actual visibility
into the state of the connection pool. With this change, we should be
able to directly observe, monitor and alert on the connection pool.
- Updated grafana version to fix a bug in the query editor that was
fixed in 8.3, even though this is a couple versions ahead of production
ref https://github.com/TryGhost/Ghost/pull/21573
- We discussed this, and figured we should get this behind the flag for
now, even though it's pulled
- This gives us the ability to do a fast revert if we have to later,
which is the point of having the labs "GA FEATURES" array
- Note this removes the labsEnabled variable in the tests to make it
clearer if the flag is off or on (it was off!)
ref https://github.com/moment/luxon/blob/master/docs/intl.md
- We noticed the following error trace: RangeError: Incorrect locale
information provided
at BatchSendingService.retryDb
(/home/ghost/node_modules/@tryghost/email-service/lib/BatchSendingService.js:639:32)
at new DateTimeFormat (<anonymous>)
at getCachedDTF
(/home/ghost/node_modules/luxon/build/node/luxon.js:621:11)
at new PolyDateFormatter
(/home/ghost/node_modules/luxon/build/node/luxon.js:842:16)
at Locale.dtFormatter
(/home/ghost/node_modules/luxon/build/node/luxon.js:1066:12)
at Formatter.dtFormatter
(/home/ghost/node_modules/luxon/build/node/luxon.js:2274:21)
at Formatter.formatDateTime
(/home/ghost/node_modules/luxon/build/node/luxon.js:2280:17)
at DateTime.toLocaleString
(/home/ghost/node_modules/luxon/build/node/luxon.js:6893:78)
at formatDateLong
(/home/ghost/node_modules/@tryghost/email-service/lib/EmailRenderer.js:45:74)
at Object.getValue
(/home/ghost/node_modules/@tryghost/email-service/lib/EmailRenderer.js:683:47)
at
/home/ghost/node_modules/@tryghost/email-service/lib/SendingService.js:158:36
at Array.map (<anonymous>)
at
/home/ghost/node_modules/@tryghost/email-service/lib/SendingService.js:154:54
at Array.map (<anonymous>)
at SendingService.buildRecipients
(/home/ghost/node_modules/@tryghost/email-service/lib/SendingService.js:151:24)
at SendingService.send
(/home/ghost/node_modules/@tryghost/email-service/lib/SendingService.js:127:33)
at response.retryDb
(/home/ghost/node_modules/@tryghost/email-service/lib/BatchSendingService.js:451:51)
- This is due to the locale being user-input - it can be set to any
string.
- In our email sending code we pass the string to luxon to format dates,
which errors if the locale is not valid according it Intl.
- This fix ensures that the locale is valid before passing it to luxon,
falling back to en-gb if the locale is not valid
no issue
- We had reintroduced nodemon in
af0f26c75f (diff-bf18f8caf848e17b35e266db04bcaeaad05a3e5d069846615d2b1260482396e1)
for the docker setup, but it has since caused some issues with the `yarn
dev` script.
- In particular, it was causing a restart while migrations were running
in development, which left a migration lock on and prevented Ghost from
starting.
- This commit removes nodemon and replaces it with node --watch, which
we had been using in the past without issues.
ref https://ghost.org/changelog/internationalization/
- It's difficult out of context to know whether to use an infinitive,
_Ajouter votre commentaire_ or an imperative, _Ajoutez votre
commentaire_. English doesn't make the difference, but French does. By
default, an infinitive is preferable because it remains a suggestion,
whereas an imperative gives an order. For example, I think that _Ajoutez
votre commentaire_ is not relevant, because you don't necessarily have
to add a comment, but you can, so you should say _Ajouter votre
commentaire_. I've tried to be as precise as possible, but you may need
to make some adjustments with the context.
- I find that the word _membre_ is too literal a translation of the
English language and includes too many innuendos such as "member of a
community", "member of a clan", "member of a club", and frankly, it's
not relevant to many blogs. The word _abonné_ is more relevant, because
it doesn't imply a social affiliation. As for the discussion, I've even
taken the liberty of writing _discussion entre lecteurs_, because that's
essentially what it is. It's not a discussion between members or
subscribers, it's a discussion between _readers_.
- Some placeholders are completely American (Jamie Larson), so I've
replaced him with a French one (Jean Martin).
- Complimentary is a word that I didn't understand in the context of
Ghost. I have the impression that it means _offert_, but I don't
understand when it's used.
refs https://linear.app/ghost/issue/AP-500
We've got a new @tryghost/activitypub package, which is gonna handle all
of the wiring between Ghost and ActivityPub. Currently that is just the
configuration of webhooks for the internal ActivityPub integration.
All this logic is run on the boot of Ghost, though notably in a
non-blocking manner, it's initialised as part of the background services
so it should not have an effect on the time to serving requests. having
said that - it needs to be defensive against errors, which is why the
entire network request is in a try/catch, as well as lookups for the
integration able to handle missing data.
Unit tests use an in-memory sqlite instance, which means we're testing a
full flow, ideally there would be a way to get the schema from Ghost for
this, but for now this is acceptable IMO.
refs https://linear.app/ghost/issue/AP-500
Rather than having to manually create an integration for communication
with the ActivityPub service, we are going to have an internal
integration which will then be used to handle webhooks between Ghost &
ActivityPub
The 'internal' type has been used to keep it out of the UI/API but
available for all Pro customers, which is necessary during the private
beta.
---------
Co-authored-by: Michael Barrett <mike@ghost.org>
This reverts commit eaed33972f.
The fix we shipped for using the `RETURN` key to create navigation items
in Admin broke something with regards to saving the URL, so we're
reverting that change.
ref https://ghost.org/changelog/internationalization/
- this fixes an error in which changing the language (or sliding the i18n
toggle) didn't correctly update email-service.
---------
Co-authored-by: Sag <guptazy@gmail.com>
ref https://linear.app/tryghost/issue/PLG-230
- adds `in_reply_to_id` to API output
- adds `in_reply_to_snippet` to API output
- dynamically generated from the HTML of the replied-to comment
- excluded if the replied-to comment has been deleted or hidden
- adds `commentSnippet` to `@tryghost/html-to-plaintext`
- skips anchor tag URLs as they won't be useful for snippet purposes
- skips blockquotes so the snippet is more likely to contain the unique content of the replied-to comment when it's quoting a previous comment
- returns a single line (no newline chars)
- allows setting `in_reply_to_id` when creating comments
- id must reference a reply with the same parent
- id must reference a published comment
- adds email notification for the original reply author when their comment is replied to
ref https://linear.app/tryghost/issue/PLG-230
- `comments.in_reply_to_id` will be used to keep a reference to the comment that the new comment was directed at
- used only for replies-to-replies, will be `null` for the top-level parent and `null` for any replies directly to that parent
- technically allows for infinite nesting within a parent comment thread but we won't be using that ability for now
- `comments.parent_id` will be kept as it provides a useful optimisation for loading the top-level comments list
- we're not using `comments.parent_id` for this to keep complexity down and avoid the need for recursive lookups
refs https://linear.app/ghost/issue/AP-500
The logic for generating identity tokens, whilst small, is something
that we don't want to duplicate - as it concerns security & access - so
can easily break interactions between different services. We're gonna
need to use identity tokens as part of the initialisation of the
activitypub service, so this is pulling it out preemptively for that use
case
We shouldn't have logic inside of the endpoint controllers anyway, so
this is kinda general cleanup.
ref
https://linear.app/ghost/issue/ENG-1746/enable-ghost-to-push-metrics-to-a-pushgateway
- Trying to get Ghost working with the prometheus pushgateway in
staging, but it's logging an error each time it tries to push the
metrics. The error output is pretty useless for debugging, so this
commit improves the error messages to make it easier to debug.
no issue
- filled some missed translations but there are yet some missed
translations to complete. Actually i'm not managing newsletters so i
need more context to translate them correctly.
ref
https://linear.app/ghost/issue/ENG-1746/enable-ghost-to-push-metrics-to-a-pushgateway
- We'd like to use prometheus to expose metrics from Ghost, but the
"standard" approach of having prometheus scrape the `/metrics` endpoint
adds some complexity and additional challenges on Pro.
- A suggested simpler alternative is to use a pushgateway, to have Ghost
_push_ metrics to prometheus, rather than have prometheus scrape the
running instances.
- This PR introduces this functionality behind a configuration.
- It also includes a refactor to the current metrics-server
implementation so all the related code for prometheus is colocated, and
the configuration is a bit more organized. `@tryghost/metrics-server`
has been renamed to `@tryghost/prometheus-metrics`, and it now includes
the metrics server and prometheus-client code itself (including the
pushgateway code)
- To enable the prometheus client alone, `prometheus:enabled` must be
true. This will _not_ enable the metrics server or the pushgateway — it
will essentially collect the metrics, but not do anything with them.
- To enable the metrics server, set `prometheus:metrics_server:enabled`
to true. You can also configure the host and port that the metrics
server should export the `/metrics` endpoint on in the
`prometheus:metrics_server` block.
- To enable the pushgateway, set `prometheus:pushgateway:enabled` to
true. You can also configure the pushgateway's `url`, the `interval` it
should push metrics in (in milliseconds) and the `jobName` in the
`prometheus:pushgateway` block.
refs https://linear.app/ghost/issue/AP-523
We want to preempt the Accept activity from our Follows, so we make the
assumption that it's succeeded. What this means is that we have to
update our `following`, `followingCount` as well as the fetched profile
to set the `isFollowing` property. This gives a more fluid user
experience when following accounts and keeps our state up to date.
Accounts where the follow request has to be accepted manually, are a
little trickier as we don't currently have easy access to the "requested
but not accepted state"
- using lodash to do this is unnecessarily heavy, so this commit
switches the code to the equivalent native version
- as mentioned in the comment I added, I think we can further optimize
this by storing it as a Set and then calling `Array.from` once, but
that's a step too far for now
- we can skip an extra `toJSON` call here by storing the value and
re-using it, which means we can cut down on the number of calls, which
can be particularly heavy when this codepath is hit hard
no issue
- following pull requests #21251 and #21497, this commit adds pt-BR
translations to newsletter.json and fixes other strings in pt-BR locale
files
ref PLG-251
- Fixed comments placeholder to change from "Start the conversation" to
"Join the discussion" when there's more than 1 comment.
- Previously, it only worked after a refresh. This fix ensures it's
reactive and would update the placeholder without the need for a
refresh.
