Overview
A Profile is a set of settings configuration that agents receive from Management Console to perform a job. You can see Agent Profiles in Management Console -> Settings -> Profile. You can also create customized profiles or edit settings in default ones.
To create a new profile click button "Add". There you can check and uncheck certain settings. Unchecked ones are inherited from default profile This means that if such a setting parameter changes in the default profile, it will automatically update that setting for any custom profile in which that parameter was left unchecked.
Checked one can be edited at any time and this change will be valid only for the current Profile where it was updated.
Agent & Job Profile Settings, and their values, are explained below (including setting name for API).
Job profile
When you create a new job, you can define the following collection of settings for that job (in “Job Profile Settings”). You can change these settings at any time by editing an existing job profile.
All agents doing the job share its group of Job Profile Settings.
The combination of Job Profile settings and Agent Profile settings make configuration complete.
|
Name |
Default value |
Purpose |
|
Event filter events_logging_filter |
<> |
Some of the events are disabled in the filter and won't show up on Management Console. Edit the filter to receive only events of interest. See here for more details. |
|
Automatically stop job after (sec) timeout
|
None |
Use this setting to force job completion in N seconds if, for some reason, the job gets stuck. Works only for Consolidation and Distribution jobs. |
|
Consoles events level events_logging_level |
Agent's level |
Deprecated. Sets the level for agent's event log, or even disables logging. If value is "Agent's level" will be same as in Agent profile. If certain level is selected, all events above it in the list are also included. See here for details about event types |
|
Create folder for pre-indexing trigger build_script_working_dir
|
Yes |
The path that was chosen for a pre-indexing trigger in a Distribution or Consolidation job will be created if its missing. |
|
Data compression tunnel_compress_stream |
No |
Enables data compression in the job to save network traffic. Has some side effect, see here for more details. |
|
File and folder name filter ignore_list |
Not set |
Global IgnoreList for all agents. See here here for details |
|
File download priority file_download_priority |
<> |
Option to prioritise active downloads based on file names. See here for more details. |
|
Follow symlinks follow_symlinks |
No |
Defines whether an Agent will follow the link and synchronise its content as files and folders. Read more about it here. |
|
Force complete job for offline Agents ignore_offline_agents |
No |
Deprecated in 3.4.0, replaced with functionality to abort a job run on error. If there are offline agents, count the job as complete for them. Otherwise, job will wait for the agents to be online. Works only for Consolidation and Distribution jobs. |
|
Force file owner to hash files
|
Yes |
It is similar to Lazy indexing, and applies to the file owner - the Agent which updated the file last. This parameter must comply with lazy indexing, see here for usage examples. Disabling it is not advisable for Synchronization jobs, especially if agents tend to go offline periodically as other agents won't be able to download the files until file owner is back online. |
|
Key token rotation policy keys_rotation_policy
|
enforced |
Introduced in v3.6.0. Increases security in Agent-to-Agent communication. For MC updated from previous version default value is "compatibility"> Read more about it here. |
|
Lazy indexing lazy_indexing
|
Agent's level |
"Agent's level" means that the value is taken from Agent Profile or is true by default if its not present in the Agent profile. Enabled Lazy indexing means that the Agent will not hash any files until some other Agent requests to download it. It improves speed of merging folder tree and getting the job into balance (especially when only few files differ on different agents). On the other hand, not having a file hash is a great risk in cases when files are renamed, new agents appear in the mesh, etc. read more about using this parameter here. Some other parameters require this option to be either on or off. |
|
Overwrite any changed files (Sync job only) overwrite_changes
|
Agent Level |
Was added to job profile in 3.8.0. "Agent's level" means that the value is taken from Agent Profile. Else, the value from Job Profile is of higher priority. |
|
Ignore changed files (Transfer jobs) ignore_modified_files
|
No |
Deprecated in 2.10.2. Replaced with "Skip file errors (transfer jobs)" in the Agent Profile Defines job run's completion in cases when some files change during transfer. If enabled and files change on source agent during transfer, the agent will skip it, give warning and continue the job run. |
|
Max file modification time (sec) scanner_max_file_age
|
Date has to be specified in unix timestamp (example: 1511136000). It limits the agent to scan and sync only files older or newer than the defined time (works for the files, added after parameter's setting up). If only the max file age is specified, then this time is relative to the current one and should be set in seconds (if file max age is set to 86400 it means that files for the last 24h will be synced only), and if min and max are specified, then this is the interval in absolute values (unix time). |
|
|
Min file modification time (sec) scanner_min_file_age
|
||
|
Network policy network_policy |
<> |
Parameter to shape connections for Agents in a job. Read more about it here. |
|
Pause job if |
<> |
Sets the rule to pause a job run for agent depending on its tag value. Read more here. |
|
Skip file errors (Transfer jobs) transfer_job_skip_
|
Agent's level |
Same as "Skip file errors" in Agents profile. Agent's level means that the value is taken from the Agent's profile per Agent. Any other value in Job profile is forced for all Agents in the job, regardless of what is set in the Agent's profile. |
|
Synchronize NTFS permissions windows_fs_acl_mode |
None |
"Don't sync Owner” will synchronize user’s SID and permissions |
|
Synchronize Posix permissions posix_fs_acl_mode |
None |
Synchronises POSIX permissions including owner and owner group: |
Agent profile
This set of settings is assigned to an agent or a group of agents. If no profile is selected, default one is applied. This profile determines settings for agents in the group. A job may be performed by a few groups of agents, each of which having their own settings.
|
Name |
Default value |
Purpose |
| CONNECTIVITY AND NETWORK SETTINGS | ||
|
Agent-Server ping messages interval (sec) ping_interval
|
300 |
The interval at which agents confirm they're connected to the Management Console. |
|
Allowed peers allowed_peers
|
<> |
Is similar to Known hosts, but accepts AgentID instead of addresses. Tracker server must be enabled and used by the Agents for the parameter to work. This parameter does not limit "Known hosts". Use it in environments with proxy server. |
|
Automatic port mapping upnp
|
Yes |
Enables or disabled automatic port mapping using UPnP. |
|
Bind interface bind_interface
|
<not set> |
Specifies network interface which will be used by Agent. Can be: |
|
Bind port bind_port
|
3839 | Specifies default listening port for Connect Agents. |
|
Custom trackers custom_tracker
|
<not set> |
Deprecated in 2.12. Use "Use tracker" parameter instead. Trackers should be entered as a single line of IP:port pairs. For multiple trackers add each on a separate line. If a single tracker has multiple IP addresses, you can add them all separated by a comma. |
|
Custom tracker modes (custom_tracker) tracker_mode
|
Load balancing |
Controls the mode which is used by the Agent if multiple trackers are defined. |
|
Known hosts folder_defaults.known_hosts
|
<not set> |
Hosts should be entered as a single line of IP:port pairs (or DNS name:port pairs*) comma-separated (no other delimiters allowed). |
|
LAN discovery mode folder_defaults.lan_discovery_mode |
Multicast and broadcast |
Setting to allow agents to send either multi- or broadcast, or neither, packets in LAN. |
|
Limit LAN bandwidth rate_limit_local_peers
|
Yes |
Applies the limit configured in the current schedule to local traffic as well as WAN traffic |
|
MTU IPv4 net.udp_ipv4_mtu
|
1300 |
Max transmission unit value. Set it lower for slow networks with high rate of packet drops |
|
MTU IPv6 net.udp_ipv6_mtu
|
1300 |
Max transmission unit value. Set it lower for slow networks with high rate of packet drops |
|
Peer queue factor peer_queue_factor
|
20 |
Integer. Factor multiplies the current bandwidth to automatically determine the size of requested data by increasing the number of requests per time unit. Set bigger value for high latency networks. Max is 20. |
|
Trackers trackers |
Default tracker |
Select the tracker that will be used by an Agent. The tracker must be configured in advance. See here for more details. |
|
Tunnel ciphers tunnel_protocols
|
DHE-PSK-AES128-GCM-SHA256;DHE-PSK-AES256-GCM-SHA384 |
Common encryption algorithms for the Agents. Read more about it here. |
|
Tunnel protocols tunnel_protocols
|
tcp; |
Starting with 3.4.0 for new installations only TCP, UTP3 (ZGT) and proxy protocols are selected by default. Sets the protocols for peer discovery and data transfer. Starting with v2.11 utp3 is the only default protocol, others need to be enabled manually. |
|
Use bind interface only use_only_bind_interface
|
No |
Forces an Agent to only connect to the interface specified in the "Bind interface" parameter. If no such interface is found, the Agent will not bind to a default interface and will stay unconnected. |
|
Use proxy server use_proxy_server |
No |
Allows Agents to send their traffic through a proxy server. |
|
Use tracker folder_defaults. |
Yes |
Parameter changed in 2.12 and shall be used to configure trackers. Enables/disables peer discovery using the tracker server |
|
WAN optimization in LAN prefer_utp2_lan
|
Yes |
Enables WAN optimization, including ZGT, for local transfers as well. |
|
FILES SETTINGS |
||
|
Allow to copy local files file_deduplication |
Off
|
If enabled, agent will copy local file pieces with matching hash instead of downloading them from remote agent. Read more here. |
|
Delay processing file system events file_delay |
See here |
Configurable delay for the Agent to process the changes in files after receiving a system notification about it. File delay configured in the Profile are of higher priority than the located locally on the Agent's disk. See here for more details. |
|
Delete files absent on source agent (Transfer jobs) folder_defaults.delete_ |
No |
Forces destination agents to delete local files from the share, if these files are absent on source agent. Works only Distribution and Consolidation jobs. |
|
Disable file deletion through context menu disable_remove_from_all_devices
|
Yes |
Deprecated starting with v3.1.0. (though it can be added as a custom parameter) If enabled, context menu "Remove from all devices" on Agents' computes will be disabled for legacy Selective Sync shares. |
|
Download priority download_priority |
No priority |
The Agent that receives the files may prioritize them by timestamp. Also it's limited by capacity of download queue - which is currently 50,000 items, i.e. Agent selects max 50,000 files to download and sorts them by given priority. |
|
Encrypt on LAN lan_encrypt_data
|
Yes |
Enables encryption for all Connect Agent data flowing in LAN |
|
Ignore symlinks ignore_symlinks
|
No |
Agents will skip symbolic links during transfer, these won't be synced. Junctions and symlinks don't sync in a Sync job to prevent filename Conflicts. Read more here. |
|
Locked files recheck interval locked_files_interval
|
600 |
Refers to local locked files on the Agent. It is the frequency at which the Agent will come back and see if the file is still locked (in cases when it's not notified by the system). Default value is 600. Put there 60 (=1 minute) if it fits your use case. Don't put a too frequent interval as the Agent will be only busy with rechecking the file. Don't put a too rare interval either as some delays in synchronization will be observed. |
|
Notify Agent of file сollision threshold notify_file_edit_conflict_time_diff
|
0 |
Option to detect that file is being edited on several Agents at the same time. Read more about it here. |
|
Overwrite any changed files (Sync job only) overwrite_changes
|
Yes |
Applies to Read-only agents in Sync jobs. Important!!! If the data is present on Read-Only agent's side (pre-seeded folder) before the job is created, the usage of this parameter with non-default value (Overwrite any changed files = NO) requires some prerequisites. Please contact support to accomplish it. |
|
Recreate deleted placeholders recreate_placeholders_on_removal
|
Yes |
Deprecated. Applied to legacy Selective Sync folders. If enabled, and a person deletes .rsl placeholder, deletion is not propagated, but instead the placeholder is recreated again. |
|
Resolve filename system conflicts fix_conflicting_paths
|
Yes (starting with v3.1.5) |
In version 3.1.5 it's Yes by default for new installations. For updated MCs previous value is preserved for the default Agent profile and other profiles inherited from it. It controls creation of filename conflicts. If disabled, agent will not check filename so as to verify whether there’s a filename conflict. It’s already disabled for transfer jobs. It greatly reduces destination I/O operations. Side effect: if there is actually a file name conflict, agent won’t create .Conflict on disk, will give an error that it cannot sync a file and the job will stuck. If parameter is enabled, the Agent will append .Conflict to the file and continue the job. Enable the parameter before creating the job. Don't enable it while job is running. |
|
Skip file errors (Transfer jobs) transfer_job_skip_ |
Yes |
If set to "Yes", files will be skipped and not transferred, (this works for transfer jobs only). Agent won't report an error, but will record event in job logs and not attempt to transfer them again until the job is stopped and launched again. If set to "No" the job will be aborted if a file error is raised on the source, the source agent will report this error on MC. The job will continue working and wait for error to be resolved if it was raised on a destination agent. |
|
Synchronize extended file attributes sync_extended_attributes
|
No |
Applies to OSX only, its com.apple.ResourceFork and alike xattrs. If disabled, it greatly reduces disk I/O. |
|
FOLDER SETTINGS |
||
|
Advanced selective sync system_native_placeholders |
True/False |
Removed from profile and enforced starting with v3.1.0. |
|
Content cleanup timeout (Advanced selective sync) (sec) transfer_job_unused_files_cleanup_timeout
|
0 |
Time period, in seconds, after which downloaded files will be dehydrated. Zero stands for never. Read more about Transparent Selective Sync on Windows and macOS |
|
Data retention (when revoking access privileges) extra_folders
|
Keep folder data on Agent |
Defines what will happen with files after admin removes job from agent's group or from Management Console. Works for Sync jobs only. |
|
Max archive file size (MB) max_file_size_for_versioning
|
1000 |
Max size of a single file for its versioning. Deleted files are always put into the Archive regardless of this size limit. |
|
Max archive file age (days) sync_trash_ttl
|
30 |
Time period for which a file will be kept in Archive. After that will be removed from Agent’s computer for good. If set as 0 (zero), files will never be deleted from Archive. |
|
Rescan interval (sec) folder_rescan_interval
|
86400 |
This option sets the folder rescan interval for changes possibly missed with OS file notifications |
|
Use archive use_synctrash_folder
|
Yes |
Enables/disables storing deleted files and file versions in Archive. |
|
LOGGING AND STATISTICS |
||
|
Agent resources update interval (sec) agent_resources_ |
60 |
The interval in seconds when agents send their current resources status to the Management Console (CPU, RAM). The larger the interval, the less frequently the Management Console is loaded. |
|
Agent status update interval (sec) agent_status_update_interval |
0 |
The interval in seconds when agents send their current transfer status to MC (% synced, count of files, etc). The larger is the interval, the lesser MC is loaded. Zero means detecting interval automatically depending on number of connected agents. |
|
Console events level (deprecated) events_logging_level
|
All |
Deprecated Sets the detail level for agent's event log which is sent to Management Console UI, or disabled showing the events. All events of the set level are additionally filtered by Events filter. See here for details about events types. |
|
Debug logging logger_mask |
Yes |
Enables or disables debug logs for agent. Debug logs are necessary to troubleshoot any issues with agents' performance |
|
Event filter events_logging_filter |
<> |
Some of the events are disabled in the filter and won't show up on Management Console. Edit the filter to receive only events of interest. See here for more details. |
|
Event logs max age (days) history_time_limit
|
30 | Number of days during which an agent keeps prior events logged in the log file |
|
DISK USAGE SETTINGS |
||
|
Advanced preallocate advanced_preallocate
|
No |
Allows Agent to write data faster to storage. Agent shall run as Admin (in Windows workstations) or System (both workstations and servers). |
|
Asynchronous disk I/O async_io
|
No |
Default is "No" for new installations starting with v2.11. DO NOT enable it for Agents installed on ARM CPU! Means reading file in several threads, optimal for local drives, especially HDD. Enable when there’s latency in reading from disk from non-local drives (slow smb) for seeding agents. Always synchronous for Mac agents. |
|
Disable differential sync prefer_net_over_disk_operations
|
Yes |
Instead of syncing only changed pieces, the whole file is fully re-synced. Set it back to False if network is weaker and slower than disk, or is expensive. |
|
Free disk space threshold (%) disk_min_free_space |
5 |
An offset to disk storage space at reaching which the agent will stop synchronization and will report the error about lack of free space on the disk. Read more about it here. |
|
Free disk space threshold (Gb) disk_min_free_space_gb |
0 |
|
|
Max size of file data chunk for network request max_chunk_size
|
0 |
Before v3.1.0 default value is 1048576. If it equals the size of torrent block, a piece will not be split into chunks which will also speed up transfer. Decrease it to the same value as torrent block size for slow networks if there are a lot of peers. If put zero here, the full piece will be transferred, not split in chunks. |
|
Min size of torrent block torrent_min_piece_size
|
1048576 |
Value must be the power of 2. |
|
Number of disk I/O threads disk_worker_pool_size
|
4 |
Number of threads each disk uses for disk operations. Works for high-latency disks with multiple worker threads support, high speed drives (fast ssd, raid). For extra slow or ultra fast drives it may be optimal to reduce the number of threads. |
|
Read files ID from disk fs_query_file_id
|
No |
Greatly improves periodic folder scan, agent won't query files' attributes data. As a side effect, a renamed file will be re-downloaded by other agents instead of performing local rename. It's forcibly enabled it for Transparent Selective Sync shares, and it's advisable ot enable it for legacy Selective Sync shares. |
|
Unbuffered disk I/O unbuffered_io
|
No | When option disabled, by setting to “No”, the Agent will use the operating system cache for synced files. |
|
Agent UI |
||
|
Allow pause from Agent UI allow_pause_from_ui
|
Yes | If enabled, a user can set agent to paused state using controls in the Agent UI. |
|
Enable bandwidth scheduler in Agent UI mc_can_manage_bandwidth_limit
|
No | If enabled, a new tab appears in Agent UI where the user can limit speed of synchronization or also pause the jobs. Read more here. |
Cloud storage profile
It's advisable to use this profile for the Agent that will be syncing data to or from a cloud storage. The profile is inherited from the default Agent profile with the below mentioned parameters changed.
| Number of disk I/O threads | 20 | Number of threads each disk uses for disk operations. |
| Rescan interval (sec) | 3600 | This option sets the folder rescan interval for changes possibly missed with OS file notifications |
A Job profile with name also exists and can be used in per-job configuration.
| ignore_symlinks | True | Since cloud storages don't interpret symbolic links, a job may stuck if there are such in the folder. This parameter instructs the cloud agent to skip the link. Error will be reported. Change it to "False" manually if using Resilio Active Everywhere v4.0.0. Synchronization if symlinks is required for Agent to be able to keep files' metadata in a cloud storage. |
|
Optimize for low bandwidth |
||
|
torrent_min_piece_size |
32768 |
In bytes. The size of piece of file for transfer. Of small, agent will read pieces from disk frequently, but it will be easier to push small pieces through limited bandwidth. |
|
max_chunk_size |
32768 |
In bytes. Chunks of file pieces. Same behavior as with file pieces |
|
peer_min_queue |
65536 |
In bytes. It's the min size of requests sent from one peer to another, within one file. the bigger it, the better it saturates bandwidth. |
|
peer_queue_factor |
2 |
It’s a factor, multiplies the current bandwidth to automatically determine the size of requested data by increasing the number of requests per time unit. Set bigger value for high latency networks. Max is 10. |
| transfer_peers_limit
|
15 |
Added in 4.1.1 |
|
Optimize for file transfer reduction |
||
|
file_deduplication |
true |
Enables copying local file pieces, read more about it here. |
|
prefer_net_over_disk_operations |
false |
"Disable differential sync" from the Agent profile. |
|
direct_torrent_enabled |
false |
Forces file to be broken into pieces. IT helps to find more of changes pieces and transfer only these instead of the whole file. |
|
lazy_indexing |
false |
Enabled lazy_indexing means that the Agent will not hash any files until some other Agent requests to download it. It improves speed of merging folder tree and getting the job into balance (especially when only few files differ on different agents). On the other hand, not having a file hash is a great risk in cases when files are renamed, new agents appear in the mesh, etc. |
|
Optimize for many Agents |
||
|
agent_status_update_interval |
60 |
Agent status update interval |
|
events_logging_level |
2 |
Console events level. 2 is for warnings only |
| agent_stats_update_interval |
120 |
Added in 4.1.1. |
| transfer_peers_limit |
15 |
Added in 4.1.1. |
| Optimize for over 10M files | ||
|
events_logging_level |
2 |
Console events level. 2 is for warnings only |
|
connect_more_peers_interval |
60 |
It defines how frequently agents create queue for download. |
| transfer_events_logging_level |
0 |
Added in 4.1.1. |
|
update_peer_stat_interval |
120 |
Deprecated in 4.1.1 |
|
state_notify_interval |
120 |
Deprecated in 4.1.1 |
|
fix_conflicting_paths |
true |
Deprecated in 4.1.1 |
|
file_deduplication |
false |
Deprecated in 4.1.1 Disables copying local file pieces, read more about it here. |
|
net_default_timeout |
300 |
Deprecated in 4.1.1 |
| Optimize for collaboration | ||
|
notify_file_edit_conflict_time_diff |
300 |
Configure a notification about same file being edited at the same time on different Agents. Is not supported for TSS shares on Windows. Read more here. |
|
fs_query_file_id |
true |
Greatly improves periodic folder scan, agent won't query files' attributes data. As a side effect, a renamed file will be re-downloaded by other agents instead of performing local rename. |
|
priority_peers |
<> |
Need to put AgentID as seen in AgentID column in MC. Read more about it here. |
|
lazy_indexing |
false |
Deprecated in 4.1.1 |
| fix_conflicting_paths |
true |
It controls creation of filename conflicts. If disabled, agent will not check filename so as to verify whether there’s a filename conflict. |
|
Optimize for VDI |
||
|
recheck_locked_files_interval |
60 |
Was introduced in 2.12.5. Read more about configuring Resilio Connect for VDI usecase here |
|
torrent_rolling_checksum |
false |
|
|
prefer_net_over_disk_operations |
true |
|
|
download_priority |
2 |
|
|
max_file_size_for_versioning |
100 |
Max size (in MB) of a file to store its version |
|
Delay processing file events |
"*.vdi": 300, "*.VDI": 300, "*.vhd*": 300, "*.VHD*": 300, "*.vbox": 300, "*.vmdk": 300, "*.VMDK": 300 |
Delay processing events from file system about file updates. |
|
ignore_list |
Pending |
Available in preset in Job Profile. |
| Optimize for DFSR replacement | ||
|
events_logging_level |
1 |
Deprecated. Sets the level for agent's event log, or even disables logging. If value is "Agent's level" will be same as in Agent profile. If certain level is selected, all events above it in the list are also included. See here for details about event types
|
|
events_logging_filter |
"0,0,3,63,3;
|
Some of the events are disabled in the filter and won't show up on Management Console. Edit the filter to receive only events of interest. See here for more details. |
|
prefer_net_over_disk_operations |
false |
Deprecated in 4.1.1 |
|
fix_conflicting_paths |
true |
It controls creation of filename conflicts. If disabled, agent will not check filename so as to verify whether there’s a filename conflict. It’s already disabled for transfer jobs. It greatly reduces destination I/O operations. Side effect: if there is actually a file name conflict, agent won’t create .Conflict on disk, will give an error that it cannot sync a file and the job will stuck. |
|
prioritize_initial_indexing_mode |
15 |
Deprecated in 4.1.1 |
|
lazy_indexing |
false |
Deprecated in 4.1.1 |
| net_default_timeout |
300 |
Timeout for Agents' connections. |
| recheck_locked_files_interval |
60 |
It is the frequency at which the Agent will come back and see if the file is still locked (in cases when it's not notified by the system). |
| Optimize for Sharepoint | ||
| disk_worker_pool_size |
5 |
Number of threads each disk uses for disk operations. Works for high-latency disks with multiple worker threads support, high speed drives (fast ssd, raid). For extra slow or ultra fast drives it may be optimal to reduce the number of threads. |
| torrent_inactive_timeout |
600 |
A timeout to unload torrent when it's there's no activity on it. |
| max_active_downloads |
50 |
Max number of files that can be uploaded/downloaded at the same time
|
| max_active_uploads |
50 |
|
| mutex_file_check_interval |
300 |
The interval for Agent to receck its ID service file |
| fs_scan_jobs_limit |
5 |
|
| prioritize_initial_indexing_mode |
3 |
The value is bitmask and sets priority for building file indexes before syncing. By default agent prioritizes indexing only for initial rescan and only when lazy_indexing is enabled. It's possible to enable priority for every rescan after restart and ignore lazy indexing state (but if it turned off, time spent for periodic scan can increase significantly) |
|
Optimize for many Agents on limited bandwidth Deprecated in 4.1.1 |
||
|
choke_peers |
true |
If enabled, starts 'muting' some agents and not sync from them, e.g. when speed is less than around 1kb/s per peer. |
|
transfer_peers_limit |
15 |
Number of agents connected to each other at a time |
|
Optimize for a network share with many files Deprecated in 4.1.1 |
||
|
lazy_indexing |
false |
Enabled lazy_indexing means that the Agent will not hash any files until some other Agent requests to download it. It improves speed of merging folder tree and getting the job into balance (especially when only few files differ on different agents). On the other hand, not having a file hash is a great risk in cases when files are renamed, new agents appear in the mesh, etc. |
|
Optimize for VPN |
||
|
prefer_utp2_lan |
TRUE |
Enables WAN optimization in LAN. |
|
net.udp_ipv4_mtu |
1300 |
Sets MTU |
|
net.udp_ipv6_mtu |
1300 |
Sets MTU |
|
Optimize for sync job: enable renames, extended attributes, conflict path resolution |
||
|
fix_conflicting_paths |
TRUE |
"Resolve filename system conflicts" from the Agent profile, see on the Agent profile tab. |
|
sync_extended_attributes |
TRUE |
"Synchronize extended file attributes" from the Agent profile |
|
fs_query_file_id |
TRUE |
"Read files ID from disk" from the Agent profile. |
|
Optimize for Cloud storage Agent |
||
|
folder_rescan_interval |
3600 |
This option decreases the default folder rescan interval since file system notifications are not supported for cloud storages. |
|
transfer_job_skip_locked_files |
false |
If set to "Yes", files will be skipped and not transferred, (this works for transfer jobs only). Agent won't report an error, but will record event in job logs and not attempt to transfer them again until the job is stopped and launched again. For skipping to work, the additional parameter lazy_indexing must also be set to 'false' in Job Profile. |
|
fix_conflicting_paths |
false |
It controls creation of filename conflicts. If disabled, agent will not check filename so as to verify whether there’s a filename conflict. |
|
disk_worker_pool_size |
20 |
Number of threads each disk uses for disk operations. Works for high-latency disks with multiple worker threads support, high speed drives (fast ssd, raid). For extra slow or ultra fast drives it may be optimal to reduce the number of threads. |
|
Optimize for Transparent Selective Sync |
||
|
lazy_indexing |
false |
Enabled lazy_indexing means that the Agent will not hash any files until some other Agent requests to download it. It improves speed of merging folder tree and getting the job into balance (especially when only few files differ on different agents). On the other hand, not having a file hash is a great risk in cases when files are renamed, new agents appear in the mesh, etc. |
|
prefer_net_over_disk_operations |
false |
"Disable differential sync" from the Agent profile. |
|
fs_query_file_id |
yes |
Greatly improves periodic folder scan, agent won't query files' attributes data. As a side effect, a renamed file will be re-downloaded by other agents instead of performing local rename. |
|
Optimize for Read-Only Agents |
||
|
enable_file_system_notifications |
false |
Read-only Agents don't need to process file system notifications since they don't upload any file changes to other Agents. |