diff --git a/documentation/en/best-practices.md b/documentation/en/best-practices.md index 0a4526c81..2243e794a 100644 --- a/documentation/en/best-practices.md +++ b/documentation/en/best-practices.md @@ -10,5 +10,5 @@ description: Curio best practices 4. Multiple workers should be started with `--post` layer to allow fast wdPost and winPost turn around time. 5. We recommend running 1 GUI layer enabled node. A cluster wide GUI can be access via this node without putting any additional strain of read operations on the DB. 6. The unsealed and sealed copies should not be stored in the same storage location. Curio will allow automatic regeneration of sealed and unsealed in future if one is lost. -7. It is recommended to create a distinct layer for each market adapter, corresponding to each minerID. This configuration enables precise control, allowing for the assignment of specific minerIDs to either the Snap Deals pipeline or the PoRep pipeline. -8. It is advised to run the market adapter on the same node that will execute the TreeD task for the PoRep pipeline or the Encode task for the Snap Deals pipeline. +7. It is recommended to create a distinct layer for each market adapter (Deprecated), corresponding to each minerID. This configuration enables precise control, allowing for the assignment of specific minerIDs to either the Snap Deals pipeline or the PoRep pipeline. +8. It is advised to run the market adapter (Deprecated) on the same node that will execute the TreeD task for the PoRep pipeline or the Encode task for the Snap Deals pipeline. diff --git a/documentation/en/curio-market/migrating-from-boost.md b/documentation/en/curio-market/migrating-from-boost.md index 3fa602e01..7398df495 100644 --- a/documentation/en/curio-market/migrating-from-boost.md +++ b/documentation/en/curio-market/migrating-from-boost.md @@ -9,8 +9,7 @@ Migrating from Boost to Curio involves transitioning all deal data, including Bo ## Build the `migrate-curio` CLI Tool * First, you need to build the `migrate-curio` tool, which is part of the Boost code base. This tool will handle the migration of deals. -* Run the following command to build the tool.\ - +* Run the following command to build the tool.\\ ```bash git checkout feat/curio-migration @@ -19,6 +18,10 @@ Migrating from Boost to Curio involves transitioning all deal data, including Bo ## Preparation for Migration +{% hint style="warning" %} +Please read the [Curio market documentation](storage-market.md) carefully before proceeding. Curio market should be fully [configured](storage-market.md#enabling-storage-market) before proceeding with migration. +{% endhint %} + * Ensure that both Boost and Curio setups are ready for the migration. **Boost must be shutdown and no deal should be in process**. You will need the following details: * Path to the Boost repository (default is `~/.boost`) * Backup your Boost repository @@ -90,14 +93,12 @@ For all deals that have been migrated and are in unsealed sectors, Curio handles ## Cleanup of Local Index Directory (LID) (optional) -* After the migration, you can clean up the old LevelDB (`LID`) data that was used by Boost:\ - +* After the migration, you can clean up the old LevelDB (`LID`) data that was used by Boost:\\ ```bash migrate-curio --boost-repo cleanup leveldb ``` -* If the `LID` store was using YugabyteDB for storing indexes, use the following command\ - +* If the `LID` store was using YugabyteDB for storing indexes, use the following command\\ ```bash ./migrate-curio cleanup yugabyte \ diff --git a/documentation/en/curio-market/storage-market.md b/documentation/en/curio-market/storage-market.md index fd679adaa..0fdbd4b0f 100644 --- a/documentation/en/curio-market/storage-market.md +++ b/documentation/en/curio-market/storage-market.md @@ -117,6 +117,21 @@ To enable the Curio market on a Curio node, the following configuration changes * `MaxDealsPerPublishMsg` for the maximum number of deals per message. * `MaxPublishDealFee` to set the fee limit for publishing deals. * If handling offline deals, configure `PieceLocator` to specify the endpoints for piece retrieval. +8. Verify that HTTP server is working: + + * Curl to your domain name and verify that server is reachable from outside\ + + + ```shell + curl https:// + + Hello, World! + -Curio + ``` + +{% hint style="warning" %} +If you do not get above output then something went wrong with configuration and you should not proceed with migration from Boost or Deal making. +{% endhint %} By applying these changes, the Curio market subsystem will be activated on the specified node(s), enabling storage deals, IPNI synchronization, and retrieval functionality. @@ -241,10 +256,11 @@ curio market seal --actor Consequences: Sealing early can speed up the process, but it may result in inefficiencies if all deals are not batched correctly. ## Offline Verified DDO deals -Curio only supports offline verified DDO deals as of now. The allocation must be created by the client for the piece and handed over to the SP alongside the data. +Curio only supports offline verified DDO deals as of now. The allocation must be created by the client for the piece and handed over to the SP alongside the data. ### How to create allocation + Clients can create allocation using the `sptool toolbox` or other methods. ```shell @@ -252,10 +268,11 @@ sptool --actor t01000 toolbox mk12-client allocate -p --piece-cid ``` -Since this is an offline deal, user must either make the data available via PieceLocator or add a data URL for this offline deal. \ No newline at end of file +Since this is an offline deal, user must either make the data available via PieceLocator or add a data URL for this offline deal.