rippled Server Doesn't Sync
This page explains possible reasons a
rippled server may start successfully, but get stuck in a "connected" state without ever fully connecting to the network. (If the server crashes during or shortly after startup, see Server Won't Start instead.)
These instructions assume you have installed
rippled on a supported platform.
Normal Syncing Behavior
Syncing with the network normally takes about 5 to 15 minutes. During that time, the server does several things:
- Loads a recommended validator list (typically from
vl.ripple.com) to determine which validators it trusts.
- Discovers peer servers and connects to them.
- Downloads the header and full state information of the latest ledger from its peers, and uses that to build its internal database of ledger data.
- Listens to its trusted validators to find which ledger hashes have been recently validated.
- Collects newly-broadcast transactions and attempts to apply them to its in-progress ledger.
If the server is unable to keep up with the network while doing these tasks, the server does not sync to the network.
First Step: Restart
Many syncing issues can be resolved by restarting the server. No matter why it didn't sync the first time, it may succeed on the second try.
If the server_info method shows a
server_state other than
full and a
server_state_duration_us of more than
900000000 (15 minutes in microseconds), then you should shut down the
rippled service, wait a few seconds, and start it again. Optionally, restart the entire machine.
If the problem persists, check the other possibilities listed on this page. If none of them seem to apply, open an issue in the
rippled repository and add the "Syncing issue" label.
Usual Causes of Syncing Issues
The most common cause of syncing issues is not meeting the system requirements. The three most common shortfalls are:
- Slow disks. You need a consistently fast solid state disk (SSD). Cloud providers like AWS usually don't guarantee disk performance, because it depends on hardware shared with other customers.
- Insufficient RAM. The memory requirements vary depending on several factors including ones that are hard to predict like network load and how people use the XRP Ledger, so it's good to have more than the minimum system requirements.
- Poor network connection. Network requirements vary the most based on how people use the XRP Ledger, but a slow or unstable connection can make it impossible to keep up with new transactions and data added to the XRP Ledger.
If you are having trouble remaining synced, double-check that your server meets the system requirements. Depending on how you use your server, you may need to meet the higher "Recommended" requirements. If you meet the "Recommended" requirements and still cannot sync, try the other possibilities on this page.
Couldn't Load Validator List
The default configuration uses a recommended list of validators retrieved from
vl.ripple.com. This list is signed by Ripple's cryptographic key pair and has a built-in expiration date. If your server cannot download the list from
vl.ripple.com for some reason, your server does not choose a set of trusted validators and cannot determine which possible ledgers to declare as valid. (If you are connected to the testnet or another parallel network, your server uses a list of trusted validators for that network instead.)
validator_list block in the server_info method response shows the status of your validator list including its expiration date. If you have a list, but it's expired, it's possible that your server had connectivity to the validator list site before but hasn't been able to connect lately, so your current list expired while your server was unable to download a more updated list.
You can also use the validator_list_sites method to get more detailed information. If the
last_refresh_time fields are missing from the validator site objects in the response, that probably indicates that your server is having trouble connecting to the validator list site. Check your firewall configuration to make sure you're not blocking outgoing traffic on port 80 (HTTP) or 443 (HTTPS). Also check that your DNS is able to resolve the domain of your validator list site.
Not Enough Peers
If your server does not connect to enough peer servers, it may not be able to download enough data to remain synced with the network as the network continues processing new transactions. This can happen if your network connection is unreliable, or if you configure your server as a private server without adding enough reliable fixed peers.
Use the peers method to get information about your server's current peers. If you have exactly 10 or 11 peers, that may indicate that your firewall is blocking incoming peer connections. Set up port forwarding to allow more incoming connections. If your server is configured as a private server, double-check the contents and syntax of the
[ips_fixed] stanza in your config file, and add more proxies or public hubs if possible.
In rare cases, corrupt data saved in your
rippled server's internal databases could cause it to fail to sync. You can safely delete your server's databases in most circumstances as long as the server is not running. Corrupt data can be the result of a momentary hardware failure when copying or writing to disk, a more serious disk failure, a different process crashing and writing to the wrong part of the disk, or other issues.
As a test, you can temporarily change the paths to your server's databases as long as you have enough free space to re-download the current ledger and store other settings.
Note: When you change the database paths, the server does not load some saved settings, such as the server's current node key pair and peer reservations. If changing the database paths fixes your server' syncing problems, you may want to re-create some of these settings.
rippledserver if it is running.
$ sudo systemctl stop rippled
Create new empty folders to hold the fresh databases.
$ mkdir /var/lib/rippled/db_new/ $ mkdir /var/lib/rippled/db_new/nudb
Edit the config file to use the new paths. Be sure to change the
pathfield of the
[node_db]stanza and the value of the
[node_db] type=NuDB path=/var/lib/rippled/db_new/nudb [database_path] /var/lib/rippled/db_new
The recommended installation uses the config file
/etc/opt/ripple/rippled.cfgby default. Other places you can put a config file include
$HOMEis the home directory of the user running
$HOME/.local/ripple/rippled.cfg, or the current working directory from where you start
$ sudo systemctl start rippled
If the server successfully syncs using the fresh databases, you can delete the folders that hold the old databases. You may also want to check for hardware failures, especially to your disk and RAM.