From e02adbf7bf3081b653509cb0fc40958845a67ebe Mon Sep 17 00:00:00 2001 From: Michael Sproul Date: Fri, 8 Dec 2023 10:45:05 +1100 Subject: [PATCH] Update docs for v4.6.0 (#4982) * Update DB migration docs * Document VC broadcast modes * Update downgrade example (#6) * update downgrade example * Add period * Add v4.1.0 --------- Co-authored-by: chonghe <44791194+chong-he@users.noreply.github.com> --- book/src/database-migrations.md | 80 ++++++++++++++++++++------------- book/src/redundancy.md | 29 +++++++++++- 2 files changed, 77 insertions(+), 32 deletions(-) diff --git a/book/src/database-migrations.md b/book/src/database-migrations.md index ed7f7d72f52..5b7b4d49371 100644 --- a/book/src/database-migrations.md +++ b/book/src/database-migrations.md @@ -16,27 +16,24 @@ validator client or the slasher**. | Lighthouse version | Release date | Schema version | Downgrade available? | |--------------------|--------------|----------------|----------------------| -| v2.0.0 | Oct 2021 | v5 | no | -| v2.1.0 | Jan 2022 | v8 | no | -| v2.2.0 | Apr 2022 | v8 | no | -| v2.3.0 | May 2022 | v9 | yes from <= v3.3.0 | -| v2.4.0 | Jul 2022 | v9 | yes from <= v3.3.0 | -| v2.5.0 | Aug 2022 | v11 | yes | -| v3.0.0 | Aug 2022 | v11 | yes | -| v3.1.0 | Sep 2022 | v12 | yes | -| v3.2.0 | Oct 2022 | v12 | yes | -| v3.3.0 | Nov 2022 | v13 | yes | -| v3.4.0 | Jan 2023 | v13 | yes | -| v3.5.0 | Feb 2023 | v15 | yes before Capella | -| v4.0.1 | Mar 2023 | v16 | yes before Capella | +| v4.6.0 | Dec 2023 | v18 | yes before Deneb | +| v4.5.0 | Sep 2023 | v17 | yes | +| v4.4.0 | Aug 2023 | v17 | yes | +| v4.3.0 | Jul 2023 | v17 | yes | | v4.2.0 | May 2023 | v17 | yes | +| v4.1.0 | Apr 2023 | v16 | no | +| v4.0.1 | Mar 2023 | v16 | no | -> **Note**: All point releases (e.g. v2.3.1) are schema-compatible with the prior minor release -> (e.g. v2.3.0). +> **Note**: All point releases (e.g. v4.4.1) are schema-compatible with the prior minor release +> (e.g. v4.4.0). > **Note**: Support for old schemas is gradually removed from newer versions of Lighthouse. We -usually do this after a major version has been out for a while and everyone has upgraded. In this -case the above table will continue to record the deprecated schema changes for reference. +usually do this after a major version has been out for a while and everyone has upgraded. Deprecated +schema versions for previous releases are archived under +[Full list of schema versions](#full-list-of-schema-versions). If you get stuck and are unable +to upgrade a **testnet** node to the latest version, sometimes it is possible to upgrade via an +intermediate version (e.g. upgrade from v3.5.0 to v4.6.0 via v4.0.1). This is never necessary +on mainnet. ## How to apply a database downgrade @@ -44,9 +41,7 @@ To apply a downgrade you need to use the `lighthouse db migrate` command with th 1. Make sure you have a copy of the latest version of Lighthouse. This will be the version that knows about the latest schema change, and has the ability to revert it. -2. Work out the schema version you would like to downgrade to by checking the table above, or the - Lighthouse release notes. E.g. if you want to downgrade from v2.3.0, which upgraded the version - from v8 to v9, then you'll want to _downgrade_ to v8 in order to run v2.2.x or earlier. +2. Work out the schema version you would like to downgrade to by checking the table above, or the [Full list of schema versions](#full-list-of-schema-versions) below. E.g. if you want to downgrade from v4.2.0, which upgraded the version from v16 to v17, then you'll want to downgrade to v16 in order to run v4.0.1. 3. **Ensure that downgrading is feasible**. Not all schema upgrades can be reverted, and some of them are time-sensitive. The release notes will state whether a downgrade is available and whether any caveats apply to it. @@ -59,14 +54,13 @@ To apply a downgrade you need to use the `lighthouse db migrate` command with th sudo -u "$LH_USER" lighthouse db migrate --to "$VERSION" --datadir "$LH_DATADIR" --network "$NET" ``` -For example if you want to downgrade to Lighthouse v2.1 or v2.2 from v2.3 and you followed Somer -Esat's guide, you would run: +For example if you want to downgrade to Lighthouse v4.0.1 from v4.2.0 and you followed Somer Esat's guide, you would run: ``` -sudo -u lighthousebeacon lighthouse db migrate --to 8 --datadir /var/lib/lighthouse --network mainnet +sudo -u lighthousebeacon lighthouse db migrate --to 16 --datadir /var/lib/lighthouse --network mainnet ``` -Where `lighthouse` is Lighthouse v2.3.0+. After the downgrade succeeds you can then replace your +Where `lighthouse` is Lighthouse v4.2.0+. After the downgrade succeeds you can then replace your global `lighthouse` binary with the older version and start your node again. ## How to apply a database upgrade @@ -161,27 +155,27 @@ lighthouse db version --network mainnet ## How to prune historic states -Pruning historic states helps in managing the disk space used by the Lighthouse beacon node by removing old beacon -states from the freezer database. This can be especially useful when the database has accumulated a significant amount -of historic data. This command is intended for nodes synced before 4.4.1, as newly synced node no longer store +Pruning historic states helps in managing the disk space used by the Lighthouse beacon node by removing old beacon +states from the freezer database. This can be especially useful when the database has accumulated a significant amount +of historic data. This command is intended for nodes synced before 4.4.1, as newly synced node no longer store historic states by default. Here are the steps to prune historic states: 1. Before running the prune command, make sure that the Lighthouse beacon node is not running. If you are using systemd, you might stop the Lighthouse beacon node with a command like: - + ```bash sudo systemctl stop lighthousebeacon ``` 2. Use the `prune-states` command to prune the historic states. You can do a test run without the `--confirm` flag to check that the database can be pruned: - + ```bash sudo -u "$LH_USER" lighthouse db prune-states --datadir "$LH_DATADIR" --network "$NET" ``` 3. If you are ready to prune the states irreversibly, add the `--confirm` flag to commit the changes: - + ```bash sudo -u "$LH_USER" lighthouse db prune-states --confirm --datadir "$LH_DATADIR" --network "$NET" ``` @@ -189,7 +183,31 @@ Here are the steps to prune historic states: The `--confirm` flag ensures that you are aware the action is irreversible, and historic states will be permanently removed. 4. After successfully pruning the historic states, you can restart the Lighthouse beacon node: - + ```bash sudo systemctl start lighthousebeacon ``` + +## Full list of schema versions + +| Lighthouse version | Release date | Schema version | Downgrade available? | +|--------------------|--------------|----------------|-------------------------------------| +| v4.6.0 | Dec 2023 | v18 | yes before Deneb | +| v4.5.0 | Sep 2023 | v17 | yes | +| v4.4.0 | Aug 2023 | v17 | yes | +| v4.3.0 | Jul 2023 | v17 | yes | +| v4.2.0 | May 2023 | v17 | yes | +| v4.1.0 | Apr 2023 | v16 | yes before Capella using <= v4.5.0 | +| v4.0.1 | Mar 2023 | v16 | yes before Capella using <= v4.5.0 | +| v3.5.0 | Feb 2023 | v15 | yes before Capella using <= v4.5.0 | +| v3.4.0 | Jan 2023 | v13 | yes using <= 4.5.0 | +| v3.3.0 | Nov 2022 | v13 | yes using <= 4.5.0 | +| v3.2.0 | Oct 2022 | v12 | yes using <= 4.5.0 | +| v3.1.0 | Sep 2022 | v12 | yes using <= 4.5.0 | +| v3.0.0 | Aug 2022 | v11 | yes using <= 4.5.0 | +| v2.5.0 | Aug 2022 | v11 | yes using <= 4.5.0 | +| v2.4.0 | Jul 2022 | v9 | yes using <= v3.3.0 | +| v2.3.0 | May 2022 | v9 | yes using <= v3.3.0 | +| v2.2.0 | Apr 2022 | v8 | no | +| v2.1.0 | Jan 2022 | v8 | no | +| v2.0.0 | Oct 2021 | v5 | no | diff --git a/book/src/redundancy.md b/book/src/redundancy.md index 77cec325371..93529295aed 100644 --- a/book/src/redundancy.md +++ b/book/src/redundancy.md @@ -75,7 +75,34 @@ lighthouse bn \ Prior to v3.2.0 fallback beacon nodes also required the `--subscribe-all-subnets` and `--import-all-attestations` flags. These flags are no longer required as the validator client will now broadcast subscriptions to all connected beacon nodes by default. This broadcast behaviour -can be disabled using the `--disable-run-on-all` flag for `lighthouse vc`. +can be disabled using the `--broadcast none` flag for `lighthouse vc` (or `--disable-run-on-all` +[deprecated]). + +### Broadcast modes + +Since v4.6.0, the Lighthouse VC can be configured to broadcast messages to all configured beacon +nodes rather than just the first available. + +The flag to control this behaviour is `--broadcast`, which takes multiple comma-separated values +from this list: + +- `subscriptions`: Send subnet subscriptions & other control messages which keep the beacon nodes + primed and ready to process messages. It is recommended to leave this enabled. +- `attestations`: Send attestations & aggregates to all beacon nodes. This can improve + propagation of attestations throughout the network, at the cost of increased load on the beacon + nodes and increased bandwidth between the VC and the BNs. +- `blocks`: Send proposed blocks to all beacon nodes. This can improve propagation of blocks + throughout the network, at the cost of slightly increased load on the beacon nodes and increased + bandwidth between the VC and the BNs. If you are looking to improve performance in a multi-BN + setup this is the first option we would recommend enabling. +- `sync-committee`: Send sync committee signatures & aggregates to all beacon nodes. This can + improve propagation of sync committee messages with similar tradeoffs to broadcasting + attestations, although occuring less often due to the infrequency of sync committee duties. +- `none`: Disable all broadcasting. This option only has an effect when provided alone, otherwise + it is ignored. Not recommended except for expert tweakers. + +The default is `--broadcast subscriptions`. To also broadcast blocks for example, use +`--broadcast subscriptions,blocks`. ## Redundant execution nodes