Print this page
NEX-9338 improve the layout of the crash directory
Reviewed by: Sanjay Nadkarni <sanjay.nadkarni@nexenta.com>
Reviewed by: Steve Peng <steve.peng@nexenta.com>
| Split |
Close |
| Expand all |
| Collapse all |
--- old/usr/src/man/man1m/savecore.1m
+++ new/usr/src/man/man1m/savecore.1m
1 1 '\" te
2 2 .\" Copyright (c) 2004, Sun Microsystems, Inc. All Rights Reserved.
3 3 .\" Copyright (c) 1983 Regents of the University of California. All rights reserved. The Berkeley software License Agreement specifies the terms and conditions for redistribution.
4 -.\" Copyright 2013 Nexenta Systems, Inc. All Rights Reserved.
5 -.TH SAVECORE 1M "Jan 30, 2013"
4 +.\" Copyright 2017 Nexenta Systems, Inc. All Rights Reserved.
5 +.TH SAVECORE 1M "Jan 08, 2017"
6 6 .SH NAME
7 7 savecore \- save a crash dump of the operating system
8 8 .SH SYNOPSIS
9 9 .LP
10 10 .nf
11 11 \fB/usr/bin/savecore\fR [\fB-Lvd\fR] [\fB-f\fR \fIdumpfile\fR] [\fIdirectory\fR]
12 12 .fi
13 13
14 14 .SH DESCRIPTION
15 -.sp
16 15 .LP
17 16 The \fBsavecore\fR utility saves a crash dump of the kernel (assuming that one
18 17 was made) and writes a reboot message in the shutdown log. By default, it is
19 18 invoked by the \fBdumpadm\fR service each time the system boots.
20 19 .sp
21 20 .LP
22 21 Depending on the \fBdumpadm\fR(1M) configuration \fBsavecore\fR saves either
23 22 the compressed or uncompressed crash dump. The compressed crash dump is saved in
24 -the file \fIdirectory\fR\fB/vmdump.\fR\fIn\fR.
23 +the file \fIdirectory\fR/data/uuid\fB/vmdump.\fR\fIn\fR, where uuid is the OS
24 +instance UUID of the image that crashed.
25 25 \fBsavecore\fR saves the uncompressed crash dump data in the file
26 -\fIdirectory\fR\fB/vmcore.\fR\fIn\fR and the kernel's namelist in
27 -\fIdirectory\fR\fB/unix.\fR\fIn.\fR The trailing \fIn\fR in the
26 +\fIdirectory\fR/data/uuid\fB/vmcore.\fR\fIn\fR and the kernel's namelist in
27 +\fIdirectory\fR/data/uuid\fB/unix.\fR\fIn.\fR The trailing \fIn\fR in the
28 28 pathnames is replaced by a number which grows every time \fBsavecore\fR is run
29 29 in that directory.
30 30 .sp
31 31 .LP
32 32 Before writing out a crash dump, \fBsavecore\fR reads a number from the file
33 33 \fIdirectory\fR\fB/minfree\fR. This is the minimum number of kilobytes that
34 34 must remain free on the file system containing \fIdirectory\fR. If after saving
35 35 the crash dump the file system containing \fIdirectory\fR would have less free
36 36 space the number of kilobytes specified in \fBminfree\fR, the crash dump is not
37 37 saved. if the \fBminfree\fR file does not exist, \fBsavecore\fR assumes a
38 38 \fBminfree\fR value of 1 megabyte.
39 39 .sp
40 40 .LP
41 41 The \fBsavecore\fR utility also logs a reboot message using facility
42 42 \fBLOG_AUTH\fR (see \fBsyslog\fR(3C)). If the system crashed as a result of a
43 43 panic, \fBsavecore\fR logs the panic string too.
44 44 .SH OPTIONS
45 -.sp
46 45 .LP
47 46 The following options are supported:
48 47 .sp
49 48 .ne 2
50 49 .na
51 50 \fB\fB-d\fR\fR
52 51 .ad
53 52 .RS 15n
54 53 Disregard dump header valid flag. Force \fBsavecore\fR to attempt to save a
55 54 crash dump even if the header information stored on the dump device indicates
56 55 the dump has already been saved.
57 56 .RE
58 57
59 58 .sp
60 59 .ne 2
61 60 .na
62 61 \fB\fB-f\fR \fIdumpfile\fR\fR
63 62 .ad
64 63 .RS 15n
65 64 Attempt to save a crash dump from the specified file instead of from the
66 65 system's current dump device. This option may be useful if the information
67 66 stored on the dump device has been copied to an on-disk file by means of the
68 67 \fBdd\fR(1M) command.
69 68 .RE
70 69
71 70 .sp
72 71 .ne 2
73 72 .na
74 73 \fB\fB-L\fR\fR
75 74 .ad
76 75 .RS 15n
77 76 Save a crash dump of the live running Solaris system, without actually
78 77 rebooting or altering the system in any way. This option forces \fBsavecore\fR
79 78 to save a live snapshot of the system to the dump device, and then immediately
80 79 to retrieve the data and to write it out to a new set of crash dump files in
81 80 the specified directory. Live system crash dumps can only be performed if you
82 81 have configured your system to have a dedicated dump device using
83 82 \fBdumpadm\fR(1M).
84 83 .sp
85 84 \fBsavecore\fR \fB-L\fR does not suspend the system, so the contents of memory
86 85 continue to change while the dump is saved. This means that live crash dumps
87 86 are not fully self-consistent.
88 87 .RE
89 88
|
↓ open down ↓ |
34 lines elided |
↑ open up ↑ |
90 89 .sp
91 90 .ne 2
92 91 .na
93 92 \fB\fB-v\fR\fR
94 93 .ad
95 94 .RS 15n
96 95 Verbose. Enables verbose error messages from \fBsavecore\fR.
97 96 .RE
98 97
99 98 .SH OPERANDS
100 -.sp
101 99 .LP
102 100 The following operands are supported:
103 101 .sp
104 102 .ne 2
105 103 .na
106 104 \fB\fIdirectory\fR\fR
107 105 .ad
108 106 .RS 13n
109 107 Save the crash dump files to the specified directory. If \fIdirectory\fR is not
110 -specified, \fBsavecore\fR saves the crash dump files to the default
111 -\fBsavecore\fR \fIdirectory\fR, configured by \fBdumpadm\fR(1M).
108 +specified, \fBsavecore\fR saves the crash dump to the default
109 +\fBsavecore\fR \fIdirectory\fR, configured by \fBdumpadm\fR(1M), and files are
110 +stored under distinct \fB\fIdirectory\fR/data/uuid folder.
112 111 .RE
113 112
114 113 .SH FILES
115 -.sp
116 114 .ne 2
117 115 .na
118 -\fB\fIdirectory\fR\fB/vmdump.\fR\fIn\fR\fR
116 +\fB\fIdirectory\fR/data/uuid\fB/vmdump.\fR\fIn\fR\fR
119 117 .ad
120 118 .RS 29n
121 119
122 120 .RE
123 121
124 122 .sp
125 123 .ne 2
126 124 .na
127 -\fB\fIdirectory\fR\fB/vmcore.\fR\fIn\fR\fR
125 +\fB\fIdirectory\fR/data/uuid\fB/vmcore.\fR\fIn\fR\fR
128 126 .ad
129 127 .RS 29n
130 128
131 129 .RE
132 130
133 131 .sp
134 132 .ne 2
135 133 .na
136 -\fB\fIdirectory\fR\fB/unix.\fR\fIn\fR\fR
134 +\fB\fIdirectory\fR/data/uuid\fB/unix.\fR\fIn\fR\fR
137 135 .ad
138 136 .RS 29n
139 137
140 138 .RE
141 139
142 140 .sp
143 141 .ne 2
144 142 .na
145 143 \fB\fIdirectory\fR\fB/bounds\fR\fR
146 144 .ad
147 145 .RS 29n
148 146
149 147 .RE
150 148
151 149 .sp
152 150 .ne 2
153 151 .na
154 152 \fB\fIdirectory\fR\fB/minfree\fR\fR
155 153 .ad
156 154 .RS 29n
157 155
158 156 .RE
159 157
|
↓ open down ↓ |
13 lines elided |
↑ open up ↑ |
160 158 .sp
161 159 .ne 2
162 160 .na
163 161 \fB\fB/var/crash/\&`uname \fR\fB-n\fR\fB\&`\fR\fR
164 162 .ad
165 163 .RS 29n
166 164 default crash dump directory
167 165 .RE
168 166
169 167 .SH SEE ALSO
170 -.sp
171 168 .LP
172 169 \fBadb\fR(1), \fBmdb\fR(1), \fBsvcs\fR(1), \fBdd\fR(1M), \fBdumpadm\fR(1M),
173 170 \fBsvcadm\fR(1M), \fBsyslog\fR(3C), \fBattributes\fR(5), \fBsmf\fR(5)
174 171 .SH NOTES
175 -.sp
176 172 .LP
177 173 The system crash dump service is managed by the service management facility,
178 174 \fBsmf\fR(5), under the service identifier:
179 175 .sp
180 176 .in +2
181 177 .nf
182 178 svc:/system/dumpadm:default
183 179 .fi
184 180 .in -2
185 181 .sp
186 182
187 183 .sp
188 184 .LP
189 185 Administrative actions on this service, such as enabling, disabling, or
190 186 requesting restart, can be performed using \fBsvcadm\fR(1M). The service's
191 187 status can be queried using the \fBsvcs\fR(1) command.
192 188 .sp
193 189 .LP
194 190 If the dump device is also being used as a swap device, you must run
195 191 \fBsavecore\fR very soon after booting, before the swap space containing the
196 192 crash dump is overwritten by programs currently running.
|
↓ open down ↓ |
11 lines elided |
↑ open up ↑ |
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX