[GH-ISSUE #154] Chain targets OFFLINE #88

Open
opened 2026-10-26 09:44:55 +00:00 by bsakaguchi · 2 comments
Owner

Originally created by @freedomsunny on GitHub (Mar 11, 2025).
Original GitHub issue: https://github.com/deepseek-ai/3FS/issues/154

As you can see. how do i troubleshooting the problem?

Image

Originally created by @freedomsunny on GitHub (Mar 11, 2025). Original GitHub issue: https://github.com/deepseek-ai/3FS/issues/154 As you can see. how do i troubleshooting the problem? ![Image](https://github.com/user-attachments/assets/c5b96f65-f979-4e7c-8344-c7f564930dec)
Author
Owner

@haohaiwei commented on GitHub (Mar 14, 2025):

每个chain 会分配replica 数量的target,从这个图可以看出,部分target 是不可用的,先检查下所有storage服务是否正常

<!-- gh-comment-id:2723218565 --> @haohaiwei commented on GitHub (Mar 14, 2025): 每个chain 会分配replica 数量的target,从这个图可以看出,部分target 是不可用的,先检查下所有storage服务是否正常
Author
Owner

@rensilin commented on GitHub (Mar 14, 2025):

每个chain 会分配replica 数量的target,从这个图可以看出,部分target 是不可用的,先检查下所有storage服务是否正常

Image
Image
Similar problem. Nodes show CONNECTED via "list-nodes", but all chains remain UNAVAILABLE - How to diagnose storage servers?
我按照3FS Setup Guide配置,4节点,所以第7步改了参数, num_nodes, num_disks_per_node, node_id_end。
这是我生成chain table的命令。

python 3FS/deploy/data_placement/src/model/data_placement.py \
   -ql -relax -type CR --num_nodes 4 --replication_factor 3 --min_targets_per_disk 6
python 3FS/deploy/data_placement/src/setup/gen_chain_table.py \
   --chain_table_type CR --node_id_begin 10001 --node_id_end 10004 \
   --num_disks_per_node 1 --num_targets_per_disk 6 \
   --target_id_prefix 1 --chain_id_prefix 9 \
   --incidence_matrix_path output/DataPlacementModel-v_4-b_8-r_6-k_3-λ_2-lb_2-ub_2/incidence_matrix.pickle

mgmtd_main-err.log, storage_main-err.log里没有错误日志,请问应该怎么检查服务是否正常呢。

<!-- gh-comment-id:2725170672 --> @rensilin commented on GitHub (Mar 14, 2025): > 每个chain 会分配replica 数量的target,从这个图可以看出,部分target 是不可用的,先检查下所有storage服务是否正常 ![Image](https://github.com/user-attachments/assets/03857a93-ffb1-4302-86fc-6e742a94c00c) ![Image](https://github.com/user-attachments/assets/67eed490-957e-47c6-86e8-e8ed915f2484) Similar problem. Nodes show CONNECTED via "list-nodes", but all chains remain UNAVAILABLE - How to diagnose storage servers? 我按照[3FS Setup Guide](https://github.com/deepseek-ai/3FS/blob/main/deploy/README.md)配置,4节点,所以第7步改了参数, num_nodes, num_disks_per_node, node_id_end。 这是我生成chain table的命令。 ``` bash python 3FS/deploy/data_placement/src/model/data_placement.py \ -ql -relax -type CR --num_nodes 4 --replication_factor 3 --min_targets_per_disk 6 python 3FS/deploy/data_placement/src/setup/gen_chain_table.py \ --chain_table_type CR --node_id_begin 10001 --node_id_end 10004 \ --num_disks_per_node 1 --num_targets_per_disk 6 \ --target_id_prefix 1 --chain_id_prefix 9 \ --incidence_matrix_path output/DataPlacementModel-v_4-b_8-r_6-k_3-λ_2-lb_2-ub_2/incidence_matrix.pickle ``` mgmtd_main-err.log, storage_main-err.log里没有错误日志,请问应该怎么检查服务是否正常呢。
Sign in to join this conversation.
No labels
pull-request
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
deepseek-ai/3FS#88
No description provided.