Ma/mscp
1
0
Fork 0
mirror of https://github.com/upa/mscp.git synced 2026-02-04 03:24:58 +08:00
Code Issues Packages Projects Releases 1 Wiki Activity

update doc for checkpoint

Browse Source
This commit is contained in:
Ryo Nakamura
2024-02-20 20:29:55 +09:00
parent fc0ced1828
commit dfdad6bca5
3 changed files with 43 additions and 40 deletions
Download Patch File Download Diff File Expand all files Collapse all files

40
doc/mscp.1.in
View File

@@ -143,7 +143,7 @@ Specifies a checkpoint file to save the state of a failed
transfer. When transferring fails due to, for example, connection
disruption or user interrupt,
.B mscp
writes the information about remaining files and chunks to the
writes the information about the remaining files and chunks to the
specified checkpoint file.
.B \-W
option with
@@ -153,27 +153,27 @@ option with
.TP
.B \-R \fICHECKPOINT\fR
Specifies a checkpoint file to resume a transfer. When this
option with a checkpoint file is passed,
Specifies a checkpoint file to resume a transfer. When a checkpoint
file is passed,
.B mscp
loads a remote host, copy direction, and files and their chunks to be
transferred from the checkpoint file. Namely,
reads the checkpoint to load a remote host, copy direction, and files
and their chunks to be transferred. Namely,
.B mscp
can resume a past failed transfer from the checkpoint. Resumeing with
a checkpoint does not require
.I source ... target
arguments. Other options for establishing SSH connections, for
example, login_name, port number, config file, should be specified as
with the failed run. In addition, checkpoint files contain files as
relative paths. Thus, you must run
arguments. Other SSH connection options, such as port number and
config file, should be specified as with the failed run. In addition,
checkpoint files have file paths as relative paths. Thus, you must run
.B mscp
in the same working directory as the failed run. You can see contents
of a checkpoint file with
in the same working directory as the failed run. You can see the
contents of a checkpoint file with the
.B mscp \-vv \-D \-R CHECKOPOINT
command (Dry-run mode). Note that the checkpoint file is not
automatically removed after the resumed transfer ends
successfully. Users should check the return value of
.B mscp
.I \-vvv \-D \-R CHECKOPOINT
command.
and remove the checkpoint if it returns 0.
.TP
@@ -355,18 +355,18 @@ Copy a local file and a directory to /tmp at a remote host:
Save a checkpoint if transfer fails:
.nf
$ mscp -W checkpoint srcdir 10.0.0.1:dst/
$ mscp -W mscp.checkpoint many-large-files 10.0.0.1:dst/
.fi
.PP
Check remaining files and chunkes, and resume a failed transfer:
Check the remaining files and chunkes, and resume the failed transfer:
.nf
# dump a checkpoint and exit (dry-run mode)
$ mscp -vvv -D -R checkpoint
# Dump the content of a checkpoint and exit (dry-run mode)
$ mscp -vv -D -R mscp.checkpoint
# resume transferring from the checkpoint
$ mscp -R checkpoint
$ mscp -R mscp.checkpoint
.fi
.PP

41
doc/mscp.rst
View File

@@ -2,7 +2,7 @@
MSCP
====
:Date: v0.1.4-20-g19c73af
:Date: v0.1.4-24-g426c3d6
NAME
====
@@ -73,22 +73,25 @@ OPTIONS
**-W CHECKPOINT**
Specifies a checkpoint file to save the state of a failed transfer.
When transferring fails due to, for example, connection disruption or
user interrupt, **mscp** writes the information about remaining files
and chunks to the specified checkpoint file. **-W** option with
user interrupt, **mscp** writes the information about the remaining
files and chunks to the specified checkpoint file. **-W** option with
**-D** (dry-run mode) only writes a checkpoint file and exits.
**-R CHECKPOINT**
Specifies a checkpoint file to resume a transfer. When this option
with a checkpoint file is passed, **mscp** loads a remote host, copy
direction, and files and their chunks to be transferred from the
checkpoint file. Namely, **mscp** can resume a past failed transfer
from the checkpoint. Resumeing with a checkpoint does not require
*source ... target* arguments. Other options for establishing SSH
connections, for example, login_name, port number, config file,
should be specified as with the failed run. In addition, checkpoint
files contain files as relative paths. Thus, you must run **mscp** in
the same working directory as the failed run. You can see contents of
a checkpoint file with **mscp** *-vvv -D -R CHECKOPOINT* command.
Specifies a checkpoint file to resume a transfer. When a checkpoint
file is passed, **mscp** reads the checkpoint to load a remote host,
copy direction, and files and their chunks to be transferred. Namely,
**mscp** can resume a past failed transfer from the checkpoint.
Resumeing with a checkpoint does not require *source ... target*
arguments. Other SSH connection options, such as port number and
config file, should be specified as with the failed run. In addition,
checkpoint files have file paths as relative paths. Thus, you must
run **mscp** in the same working directory as the failed run. You can
see the contents of a checkpoint file with the **mscp -vv -D -R
CHECKOPOINT** command (Dry-run mode). Note that the checkpoint file
is not automatically removed after the resumed transfer ends
successfully. Users should check the return value of **mscp** and
remove the checkpoint if it returns 0.
**-s MIN_CHUNK_SIZE**
Specifies the minimum chunk size. **mscp** divides a file into chunks
@@ -224,17 +227,17 @@ Save a checkpoint if transfer fails:
::
$ mscp -W checkpoint srcdir 10.0.0.1:dst/
$ mscp -W mscp.checkpoint many-large-files 10.0.0.1:dst/
Check remaining files and chunkes, and resume a failed transfer:
Check the remaining files and chunkes, and resume the failed transfer:
::
# dump a checkpoint and exit (dry-run mode)
$ mscp -vvv -D -R checkpoint
# Dump the content of a checkpoint and exit (dry-run mode)
$ mscp -vv -D -R mscp.checkpoint
# resume transferring from the checkpoint
$ mscp -R checkpoint
$ mscp -R mscp.checkpoint
In a long fat network, following options might improve performance:

2
src/main.c
View File

@@ -42,7 +42,7 @@ void usage(bool print_help)
"(default: 8)\n"
" -I INTERVAL interval between SSH connection attempts (default: 0)\n"
" -W CHECKPOINT write states to the checkpoint if transfer fails\n"
" -R CHECKPOINT resume the transfer from the checkpoint\n"
" -R CHECKPOINT resume transferring from the checkpoint\n"
"\n"
" -s MIN_CHUNK_SIZE min chunk size (default: 64MB)\n"
" -S MAX_CHUNK_SIZE max chunk size (default: filesize/nr_conn)\n"